Improvements from TheForge (see description)

The work was performed by collaboration of TheForge and Google. I am
merely splitting it up into smaller PRs and cleaning it up.

This is the most "risky" PR so far because the previous ones have been
miscellaneous stuff aimed at either [improve
debugging](https://github.com/godotengine/godot/pull/90993) (e.g. device
lost), [improve Android
experience](https://github.com/godotengine/godot/pull/96439) (add Swappy
for better Frame Pacing + Pre-Transformed Swapchains for slightly better
performance), or harmless [ASTC
improvements](https://github.com/godotengine/godot/pull/96045) (better
performance by simply toggling a feature when available).

However this PR contains larger modifications aimed at improving
performance or reducing memory fragmentation. With greater
modifications, come greater risks of bugs or breakage.

Changes introduced by this PR:

TBDR GPUs (e.g. most of Android + iOS + M1 Apple) support rendering to
Render Targets that are not backed by actual GPU memory (everything
stays in cache). This works as long as load action isn't `LOAD`, and
store action must be `DONT_CARE`. This saves VRAM (it also makes
painfully obvious when a mistake introduces a performance regression).
Of particular usefulness is when doing MSAA and keeping the raw MSAA
content is not necessary.

Some GPUs get faster when the sampler settings are hard-coded into the
GLSL shaders (instead of being dynamically bound at runtime). This
required changes to the GLSL shaders, PSO creation routines, Descriptor
creation routines, and Descriptor binding routines.

 - `bool immutable_samplers_enabled = true`

Setting it to false enforces the old behavior. Useful for debugging bugs
and regressions.

Immutable samplers requires that the samplers stay... immutable, hence
this boolean is useful if the promise gets broken. We might want to turn
this into a `GLOBAL_DEF` setting.

Instead of creating dozen/hundreds/thousands of `VkDescriptorSet` every
frame that need to be freed individually when they are no longer needed,
they all get freed at once by resetting the whole pool. Once the whole
pool is no longer in use by the GPU, it gets reset and its memory
recycled. Descriptor sets that are created to be kept around for longer
or forever (i.e. not created and freed within the same frame) **must
not** use linear pools. There may be more than one pool per frame. How
many pools per frame Godot ends up with depends on its capacity, and
that is controlled by
`rendering/rendering_device/vulkan/max_descriptors_per_pool`.

- **Possible improvement for later:** It should be possible for Godot
to adapt to how many descriptors per pool are needed on a per-key basis
(i.e. grow their capacity like `std::vector` does) after rendering a few
frames; which would be better than the current solution of having a
single global value for all pools (`max_descriptors_per_pool`) that the
user needs to tweak.

 - `bool linear_descriptor_pools_enabled = true`

Setting it to false enforces the old behavior. Useful for debugging bugs
and regressions.
Setting it to false is required when workarounding driver bugs (e.g.
Adreno 730).

A ridiculous optimization. Ridiculous because the original code
should've done this in the first place. Previously Godot was doing the
following:

  1. Create a command buffer **pool**. One per frame.
  2. Create multiple command buffers from the pool in point 1.
3. Call `vkBeginCommandBuffer` on the cmd buffer in point 2. This
resets the cmd buffer because Godot requests the
`VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT` flag.
  4. Add commands to the cmd buffers from point 2.
  5. Submit those commands.
6. On frame N + 2, recycle the buffer pool and cmd buffers from pt 1 &
2, and repeat from step 3.

The problem here is that step 3 resets each command buffer individually.
Initially Godot used to have 1 cmd buffer per pool, thus the impact is
very low.

But not anymore (specially with Adreno workarounds to force splitting
compute dispatches into a new cmd buffer, more on this later). However
Godot keeps around a very low amount of command buffers per frame.

The recommended method is to reset the whole pool, to reset all cmd
buffers at once. Hence the new steps would be:

  1. Create a command buffer **pool**. One per frame.
  2. Create multiple command buffers from the pool in point 1.
3. Call `vkBeginCommandBuffer` on the cmd buffer in point 2, which is
already reset/empty (see step 6).
  4. Add commands to the cmd buffers from point 2.
  5. Submit those commands.
6. On frame N + 2, recycle the buffer pool and cmd buffers from pt 1 &
2, call `vkResetCommandPool` and repeat from step 3.

**Possible issues:** @dariosamo added `transfer_worker` which creates a
command buffer pool:

```cpp
transfer_worker->command_pool =
driver->command_pool_create(transfer_queue_family,
RDD::COMMAND_BUFFER_TYPE_PRIMARY);
```

As expected, validation was complaining that command buffers were being
reused without being reset (that's good, we now know Validation Layers
will warn us of wrong use).
I fixed it by adding:

```cpp
void RenderingDevice::_wait_for_transfer_worker(TransferWorker
*p_transfer_worker) {
	driver->fence_wait(p_transfer_worker->command_fence);
	driver->command_pool_reset(p_transfer_worker->command_pool); //
! New line !
```

**Secondary cmd buffers are subject to the same issue but I didn't alter
them. I talked this with Dario and he is aware of this.**
Secondary cmd buffers are currently disabled due to other issues (it's
disabled on master).

 - `bool RenderingDeviceCommons::command_pool_reset_enabled`

Setting it to false enforces the old behavior. Useful for debugging bugs
and regressions.

There's no other reason for this boolean. Possibly once it becomes well
tested, the boolean could be removed entirely.

Adds `command_bind_render_uniform_sets` and
`add_draw_list_bind_uniform_sets` (+ compute variants).

It performs the same as `add_draw_list_bind_uniform_set` (notice
singular vs plural), but on multiple consecutive uniform sets, thus
reducing graph and draw call overhead.

 - `bool descriptor_set_batching = true;`

Setting it to false enforces the old behavior. Useful for debugging bugs
and regressions.

There's no other reason for this boolean. Possibly once it becomes well
tested, the boolean could be removed entirely.

Godot currently does the following:

 1. Fill the entire cmd buffer with commands.
 2. `submit()`
    - Wait with a semaphore for the swapchain.
- Trigger a semaphore to indicate when we're done (so the swapchain
can submit).
 3. `present()`

The optimization opportunity here is that 95% of Godot's rendering is
done offscreen.
Then a fullscreen pass copies everything to the swapchain. Godot doesn't
practically render directly to the swapchain.

The problem with this is that the GPU has to wait for the swapchain to
be released **to start anything**, when we could start *much earlier*.
Only the final blit pass must wait for the swapchain.

TheForge changed it to the following (more complicated, I'm simplifying
the idea):

 1. Fill the entire cmd buffer with commands.
 2. In `screen_prepare_for_drawing` do `submit()`
    - There are no semaphore waits for the swapchain.
    - Trigger a semaphore to indicate when we're done.
3. Fill a new cmd buffer that only does the final blit to the
swapchain.
 4. `submit()`
    - Wait with a semaphore for the submit() from step 2.
- Wait with a semaphore for the swapchain (so the swapchain can
submit).
- Trigger a semaphore to indicate when we're done (so the swapchain
can submit).
 5. `present()`

Dario discovered this problem independently while working on a different
platform.

**However TheForge's solution had to be rewritten from scratch:** The
complexity to achieve the solution was high and quite difficult to
maintain with the way Godot works now (after Übershaders PR).
But on the other hand, re-implementing the solution became much simpler
because Dario already had to do something similar: To fix an Adreno 730
driver bug, he had to implement splitting command buffers. **This is
exactly what we need!**. Thus it was re-written using this existing
functionality for a new purpose.

To achieve this, I added a new argument, `bool p_split_cmd_buffer`, to
`RenderingDeviceGraph::add_draw_list_begin`, which is only set to true
by `RenderingDevice::draw_list_begin_for_screen`.

The graph will split the draw list into its own command buffer.

 - `bool split_swapchain_into_its_own_cmd_buffer = true;`

Setting it to false enforces the old behavior. This might be necessary
for consoles which follow an alternate solution to the same problem.
If not, then we should consider removing it.

PR #90993 added `shader_destroy_modules()` but it was not actually in
use.

This PR adds several places where `shader_destroy_modules()` is called
after initialization to free up memory of SPIR-V structures that are no
longer needed.

servers/rendering/rendering_device.h

@@ -191,7 +191,12 @@ private:
 	Error _buffer_initialize(Buffer *p_buffer, const uint8_t *p_data, size_t p_data_size, uint32_t p_required_align = 32);
 
 	void update_perf_report();
-
+	// flag for batching descriptor sets
+	bool descriptor_set_batching = true;
+	// When true, the final draw call that copies our offscreen result into the Swapchain is put into its
+	// own cmd buffer, so that the whole rendering can start early instead of having to wait for the
+	// swapchain semaphore to be signaled (which causes bubbles).
+	bool split_swapchain_into_its_own_cmd_buffer = true;
 	uint32_t gpu_copy_count = 0;
 	uint32_t copy_bytes_count = 0;
 	String perf_report_text;
@@ -543,6 +548,7 @@ public:
 	void framebuffer_set_invalidation_callback(RID p_framebuffer, InvalidationCallback p_callback, void *p_userdata);
 
 	FramebufferFormatID framebuffer_get_format(RID p_framebuffer);
+	Size2 framebuffer_get_size(RID p_framebuffer);
 
 	/*****************/
 	/**** SAMPLER ****/
@@ -843,6 +849,7 @@ public:
 	RID shader_create_from_spirv(const Vector<ShaderStageSPIRVData> &p_spirv, const String &p_shader_name = "");
 	RID shader_create_from_bytecode(const Vector<uint8_t> &p_shader_binary, RID p_placeholder = RID());
 	RID shader_create_placeholder();
+	void shader_destroy_modules(RID p_shader);
 
 	uint64_t shader_get_vertex_input_attribute_mask(RID p_shader);
 
@@ -855,13 +862,20 @@ public:
 		STORAGE_BUFFER_USAGE_DISPATCH_INDIRECT = 1,
 	};
 
+	/*****************/
+	/**** BUFFERS ****/
+	/*****************/
+
 	RID uniform_buffer_create(uint32_t p_size_bytes, const Vector<uint8_t> &p_data = Vector<uint8_t>());
 	RID storage_buffer_create(uint32_t p_size, const Vector<uint8_t> &p_data = Vector<uint8_t>(), BitField<StorageBufferUsage> p_usage = 0);
+
 	RID texture_buffer_create(uint32_t p_size_elements, DataFormat p_format, const Vector<uint8_t> &p_data = Vector<uint8_t>());
 
 	struct Uniform {
 		UniformType uniform_type = UNIFORM_TYPE_IMAGE;
 		uint32_t binding = 0; // Binding index as specified in shader.
+		// This flag specifies that this is an immutable sampler to be set when creating pipeline layout.
+		bool immutable_sampler = false;
 
 	private:
 		// In most cases only one ID is provided per binding, so avoid allocating memory unnecessarily for performance.
@@ -922,6 +936,9 @@ public:
 		_FORCE_INLINE_ Uniform() = default;
 	};
 
+	typedef Uniform PipelineImmutableSampler;
+	RID shader_create_from_bytecode_with_samplers(const Vector<uint8_t> &p_shader_binary, RID p_placeholder = RID(), const Vector<PipelineImmutableSampler> &p_immutable_samplers = Vector<PipelineImmutableSampler>());
+
 private:
 	static const uint32_t MAX_UNIFORM_SETS = 16;
 	static const uint32_t MAX_PUSH_CONSTANT_SIZE = 128;
@@ -963,11 +980,23 @@ private:
 	void _uniform_set_update_shared(UniformSet *p_uniform_set);
 
 public:
+	/** Bake a set of uniforms that can be bound at runtime with the given shader.
+	 * @remark				Setting p_linear_pool = true while keeping the RID around for longer than the current frame will result in undefined behavior.
+	 * @param p_uniforms	The uniforms to bake into a set.
+	 * @param p_shader		The shader you intend to bind these uniforms with.
+	 * @param p_set_index	The set. Should be in range [0; 4)
+	 *						The value 4 comes from physical_device_properties.limits.maxBoundDescriptorSets. Vulkan only guarantees maxBoundDescriptorSets >= 4 (== 4 is very common on Mobile).
+	 * @param p_linear_pool	If you call this function every frame (and free the returned RID within the same frame!), set it to true for better performance.
+	 *						If you plan on keeping the return value around for more than one frame (e.g. Sets that are created once and reused forever) you MUST set it to false.
+	 * @return				Baked descriptor set.
+	 */
 	template <typename Collection>
-	RID uniform_set_create(const Collection &p_uniforms, RID p_shader, uint32_t p_shader_set);
+	RID uniform_set_create(const Collection &p_uniforms, RID p_shader, uint32_t p_shader_set, bool p_linear_pool = false);
 	bool uniform_set_is_valid(RID p_uniform_set);
 	void uniform_set_set_invalidation_callback(RID p_uniform_set, InvalidationCallback p_callback, void *p_userdata);
 
+	bool uniform_sets_have_linear_pools() const;
+
 	/*******************/
 	/**** PIPELINES ****/
 	/*******************/
@@ -1181,6 +1210,7 @@ public:
 	void draw_list_draw(DrawListID p_list, bool p_use_indices, uint32_t p_instances = 1, uint32_t p_procedural_vertices = 0);
 	void draw_list_draw_indirect(DrawListID p_list, bool p_use_indices, RID p_buffer, uint32_t p_offset = 0, uint32_t p_draw_count = 1, uint32_t p_stride = 0);
 
+	void draw_list_set_viewport(DrawListID p_list, const Rect2 &p_rect);
 	void draw_list_enable_scissor(DrawListID p_list, const Rect2 &p_rect);
 	void draw_list_disable_scissor(DrawListID p_list);
 
@@ -1374,7 +1404,8 @@ private:
 		// This must have the same size of the transfer worker pool.
 		TightLocalVector<RDD::SemaphoreID> transfer_worker_semaphores;
 
-		// Extra command buffer pool used for driver workarounds.
+		// Extra command buffer pool used for driver workarounds or to reduce GPU bubbles by
+		// splitting the final render pass to the swapchain into its own cmd buffer.
 		RDG::CommandBufferPool command_buffer_pool;
 
 		struct Timestamp {
@@ -1405,8 +1436,14 @@ private:
 	uint64_t texture_memory = 0;
 	uint64_t buffer_memory = 0;
 
+protected:
+	void execute_chained_cmds(bool p_present_swap_chain,
+			RenderingDeviceDriver::FenceID p_draw_fence,
+			RenderingDeviceDriver::SemaphoreID p_dst_draw_semaphore_to_signal);
+
+public:
 	void _free_internal(RID p_id);
-	void _begin_frame();
+	void _begin_frame(bool p_presented = false);
 	void _end_frame();
 	void _execute_frame(bool p_present);
 	void _stall_for_previous_frames();