[ExecuTorch][WebGPU] Coordinate dynamic dispatch routes

[JCNTH]

Stack from ghstack (oldest at bottom):

• [ExecuTorch][WebGPU] Preserve Linear and Q4 embedding params across dynamic resize #21483
• [ExecuTorch][WebGPU] Generate logical and arithmetic binary shader variants #21451
• [ExecuTorch][WebGPU] Generate extrema and unary shader variants #21450
• [ExecuTorch][WebGPU] Generate typed to_copy shader variants #21449
• [ExecuTorch][WebGPU] Make WGSL generation rollback-safe #21448
• [ExecuTorch][WebGPU] Add Qwen3 K16 attention and fp16 KV cache #21136
• [ExecuTorch][WebGPU] Add HuggingFace rotate-half RoPE #21135
• [ExecuTorch][WebGPU] Repair dynamic SDPA routing, add attestation #21134
• [ExecuTorch][WebGPU] Select emdawnwebgpu for Emscripten builds #21133
• [ExecuTorch][WebGPU] Add Llama K16 online causal attention #21132
• [ExecuTorch][WebGPU] Replace broad QKV fusion with BK64 kernel #21131
• [ExecuTorch][WebGPU] Route exact prefill shapes to the BK64 GEMM #21130
• [ExecuTorch][WebGPU] Fuse the SwiGLU elementwise chain #21129
• -> [ExecuTorch][WebGPU] Coordinate dynamic dispatch routes #21128
• [ExecuTorch][WebGPU] Add certified scoped output suppression #21127

Centralize dynamic dispatch routing, resize hooks, and workgroup-grid ownership

Quantized-linear and attention optimizations select specialized dispatch ranges from live sequence lengths, but the original handlers duplicated resize bookkeeping and exposed dispatch indices. This change centralizes route coordination and dynamic-grid ownership in WebGPUGraph.

Key changes:

• WebGPUUtils.h — register mutually exclusive routes and validate their ranges.
• WebGPUGraph.{h,cpp} — add typed named resize hooks, descriptor-driven compute construction, graph-owned parameter buffers and dynamic grids, staged grid updates, and exception-safe retry.
• QuantizedLinear.cpp, Sdpa.cpp, and SdpaFdDecode.{h,cpp} — select specialized ranges through the shared route registry.
• test_compute_dispatch.cpp — cover invalid triggers, route overlap, rollback/retry, ownership, and repeated transitions.

The route registry is WebGPU-specific. Graph-owned grid recomputation mirrors Vulkan's DynamicDispatchNode lifecycle without introducing a Vulkan-style node hierarchy.

Co-authored-with: Claude Code.
@exported-using-ghexport

Differential Revision: D113171748

Differential Revision: D113171748

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21128

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 8bdee7c with merge base 8200b9e ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.