atomic-vec2u-min-max support in wgpu/WebGPU

[dnsn021]

Hi,

I'm working on a GPU-driven renderer with a Nanite-like compute-shader-based software rasterization / visibility-buffer pipeline, targeting both native wgpu and WebGPU/WASM.

The compute rasterizer needs to atomically select a packed (depth, primitive_id) value for each pixel. Native backends can use 64-bit atomics, but WebGPU currently has no equivalent available through wgpu, so I currently disable this compute rasterization path on the Web backend.

WGSL/WebGPU has now added atomic_vec2u_min_max / "atomic-vec2u-min-max" for this use case:

• [GPUWeb proposal](https://github.com/gpuweb/gpuweb/blob/main/proposals/atomic-64-min-max.md)
• [GPUWeb issue #5071](Add limited support for 64 bit atomics gpuweb/gpuweb#5071)
• [WGSL specification PR #5610](Spec changes for atomic vec2u gpuweb/gpuweb#5610)
• [Dawn implementation](https://dawn.googlesource.com/dawn/+/d648286b72469897d5d05714a526d3099fe6a20c)

I couldn't find a corresponding tracking issue in wgpu.

Is support for this feature already planned, or would it make sense to track it in wgpu?

[ErichDonGubler]

Just filed #10435. 🫡