[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction

[JCNTH]

Stack from ghstack (oldest at bottom):

• -> [ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21602
• [ExecuTorch][WebGPU] Cover Conv1d routes and dynamic manifests #21601
• [ExecuTorch][WebGPU] Support general groups=1 Conv1d #21600
• [ExecuTorch][WebGPU] Test byte-packed BOOL tails #21599
• [ExecuTorch][WebGPU] Support byte-packed BOOL storage #21598
• [ExecuTorch][Vulkan] Preserve persistent buffer mutations #21597
• [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

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

• GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
• Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
• to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
• Production builds exclude timestamp-query validation and symbols.
• Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

• 📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

• pull / android / run-emulator (gh)
  adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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.