Clarify the check on buffer offset in buffer-texture copy with depth …

…stencil formats (gpuweb#1528)

* Clarify the check on buffer offset in buffer-texture copy with depth stencil formats

This patch intends to clarify the validation rules on the
buffer offset in copyBufferToTexture() and copyTextureToBuffer() with
depth stencil formats. (The buffer offset must be a multiple of 4 when
the texture has a depth/stencil format).

According to Vulkan SPEC (1.2.172, chapter 19.4):
- VUID-vkCmdCopyBufferToImage-srcImage-04053
  If dstImage has a depth/stencil format, the bufferOffset member of any
  element of pRegions must be a multiple of 4
- VUID-vkCmdCopyImageToBuffer-srcImage-04053
  If srcImage has a depth/stencil format, the bufferOffset member of any
  element of pRegions must be a multiple of 4

* Small fix

spec/index.bs

@@ -5254,8 +5254,11 @@ Issue: convert "Valid Texture Copy Range" into an algorithm with parameters, sim
                         |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}}.
                         See [[#depth-formats|depth-formats]].
                 - [=Valid Texture Copy Range=] applies to |destination| and |copySize|.
-                - |source|.{{GPUImageDataLayout/offset}} is a multiple of the [=texel block size=]
-                    of |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}}.
+                - If |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}} is not a depth/stencil format:
+                    - |source|.{{GPUImageDataLayout/offset}} is a multiple of the [=texel block size=] of
+                        |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}}.
+                - If |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}} is a depth/stencil format:
+                    - |source|.{{GPUImageDataLayout/offset}} is a multiple of 4.
                 - [$validating linear texture data$](|source|,
                     |source|.{{GPUImageCopyBuffer/buffer}}.{{GPUBuffer/[[size]]}},
                     |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}},
@@ -5295,8 +5298,11 @@ Issue: convert "Valid Texture Copy Range" into an algorithm with parameters, sim
                 - |destination|.{{GPUImageCopyBuffer/buffer}}.{{GPUBuffer/[[usage]]}} contains
                     {{GPUBufferUsage/COPY_DST}}.
                 - [=Valid Texture Copy Range=] applies to |source| and |copySize|.
-                - |destination|.{{GPUImageDataLayout/offset}} is a multiple of the [=texel block size=]
-                    of |destination|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}}.
+                - If |source|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}} is not a depth/stencil format:
+                    - |destination|.{{GPUImageDataLayout/offset}} is a multiple of the [=texel block size=] of
+                        |source|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}}.
+                - If |source|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}} is a depth/stencil format:
+                    - |destination|.{{GPUImageDataLayout/offset}} is a multiple of 4.
                 - [$validating linear texture data$](|destination|,
                     |destination|.{{GPUImageCopyBuffer/buffer}}.{{GPUBuffer/[[size]]}},
                     |source|.{{GPUImageCopyTexture/texture}}.{{GPUTexture/[[format]]}},