Use instancing for sprites (#9597)

# Objective

- Supercedes #8872 
- Improve sprite rendering performance after the regression in #9236 

## Solution

- Use an instance-rate vertex buffer to store per-instance data.
- Store color, UV offset and scale, and a transform per instance.
- Convert Sprite rect, custom_size, anchor, and flip_x/_y to an affine
3x4 matrix and store the transpose of that in the per-instance data.
This is similar to how MeshUniform uses transpose affine matrices.
- Use a special index buffer that has batches of 6 indices referencing 4
vertices. The lower 2 bits indicate the x and y of a quad such that the
corners are:
  ```
  10    11

  00    01
  ```
UVs are implicit but get modified by UV offset and scale The remaining
upper bits contain the instance index.

## Benchmarks

I will compare versus `main` before #9236 because the results should be
as good as or faster than that. Running `bevymark -- 10000 16` on an M1
Max with `main` at `e8b38925` in yellow, this PR in red:

![Screenshot 2023-08-27 at 18 44
10](https://github.com/bevyengine/bevy/assets/302146/bdc5c929-d547-44bb-b519-20dce676a316)

Looking at the median frame times, that's a 37% reduction from before.

---

## Changelog

- Changed: Improved sprite rendering performance by leveraging an
instance-rate vertex buffer.

---------

Co-authored-by: Giacomo Stevanato <[email protected]>

Author: superdump

crates/bevy_pbr/src/render/mesh_functions.wgsl

@@ -4,26 +4,7 @@
 #import bevy_pbr::mesh_bindings       mesh
 #import bevy_pbr::mesh_types          MESH_FLAGS_SIGN_DETERMINANT_MODEL_3X3_BIT
 #import bevy_render::instance_index   get_instance_index
-
-fn affine_to_square(affine: mat3x4<f32>) -> mat4x4<f32> {
-    return transpose(mat4x4<f32>(
-        affine[0],
-        affine[1],
-        affine[2],
-        vec4<f32>(0.0, 0.0, 0.0, 1.0),
-    ));
-}
-
-fn mat2x4_f32_to_mat3x3_unpack(
-    a: mat2x4<f32>,
-    b: f32,
-) -> mat3x3<f32> {
-    return mat3x3<f32>(
-        a[0].xyz,
-        vec3<f32>(a[0].w, a[1].xy),
-        vec3<f32>(a[1].zw, b),
-    );
-}
+#import bevy_render::maths            affine_to_square, mat2x4_f32_to_mat3x3_unpack
 
 fn get_model_matrix(instance_index: u32) -> mat4x4<f32> {
     return affine_to_square(mesh[get_instance_index(instance_index)].model);

crates/bevy_pbr/src/render/mesh_types.wgsl

@@ -2,7 +2,7 @@
 
 struct Mesh {
     // Affine 4x3 matrices transposed to 3x4
-    // Use bevy_pbr::mesh_functions::affine_to_square to unpack
+    // Use bevy_render::maths::affine_to_square to unpack
     model: mat3x4<f32>,
     previous_model: mat3x4<f32>,
     // 3x3 matrix packed in mat2x4 and f32 as:

crates/bevy_render/src/lib.rs

@@ -234,6 +234,8 @@ pub struct RenderApp;
 
 pub const INSTANCE_INDEX_SHADER_HANDLE: HandleUntyped =
     HandleUntyped::weak_from_u64(Shader::TYPE_UUID, 10313207077636615845);
+pub const MATHS_SHADER_HANDLE: HandleUntyped =
+    HandleUntyped::weak_from_u64(Shader::TYPE_UUID, 10665356303104593376);
 
 impl Plugin for RenderPlugin {
     /// Initializes the renderer, sets up the [`RenderSet`](RenderSet) and creates the rendering sub-app.
@@ -391,6 +393,7 @@ impl Plugin for RenderPlugin {
                 "BASE_INSTANCE_WORKAROUND".into()
             ]
         );
+        load_internal_asset!(app, MATHS_SHADER_HANDLE, "maths.wgsl", Shader::from_wgsl);
         if let Some(future_renderer_resources) =
             app.world.remove_resource::<FutureRendererResources>()
         {

crates/bevy_render/src/maths.wgsl

@@ -0,0 +1,21 @@
+#define_import_path bevy_render::maths
+
+fn affine_to_square(affine: mat3x4<f32>) -> mat4x4<f32> {
+    return transpose(mat4x4<f32>(
+        affine[0],
+        affine[1],
+        affine[2],
+        vec4<f32>(0.0, 0.0, 0.0, 1.0),
+    ));
+}
+
+fn mat2x4_f32_to_mat3x3_unpack(
+    a: mat2x4<f32>,
+    b: f32,
+) -> mat3x3<f32> {
+    return mat3x3<f32>(
+        a[0].xyz,
+        vec3<f32>(a[0].w, a[1].xy),
+        vec3<f32>(a[1].zw, b),
+    );
+}

crates/bevy_render/src/render_resource/buffer_vec.rs

@@ -144,6 +144,14 @@ impl<T: Pod> BufferVec<T> {
     pub fn clear(&mut self) {
         self.values.clear();
     }
+
+    pub fn values(&self) -> &Vec<T> {
+        &self.values
+    }
+
+    pub fn values_mut(&mut self) -> &mut Vec<T> {
+        &mut self.values
+    }
 }
 
 impl<T: Pod> Extend<T> for BufferVec<T> {