[ExecuTorch][WebGPU] Dynamic resize hooks for sigmoid and select_copy

[JulianCloudNTH]

Stack from ghstack (oldest at bottom):

• [ExecuTorch][WebGPU] 2D compute dispatch tests — prefill golden + fold unit test #20584
• [ExecuTorch][WebGPU] 2D compute dispatch — lift the 65535 per-dim cap (prefill path) #20583
• [ExecuTorch][WebGPU] Dynamic-shape integration test (allocate-at-max + per-op resize) #20582
• [ExecuTorch][WebGPU] Dynamic resize hook for slice_copy (dynamic start/end) #20581
• [ExecuTorch][WebGPU] Dynamic resize hook for SDPA (live seq-len S) #20580
• [ExecuTorch][WebGPU] Dynamic resize hook for view_copy #20579
• -> [ExecuTorch][WebGPU] Dynamic resize hooks for sigmoid and select_copy #20578
• [ExecuTorch][WebGPU] Dynamic resize hooks for add and mul #20577
• [ExecuTorch][WebGPU] Dynamic resize hook for linear_q4gsw #20576
• [ExecuTorch][WebGPU] Dynamic resize hooks for rms_norm, embedding, rope #20575
• [ExecuTorch][WebGPU] Dynamic tensor-shape resize engine core #20574
• [ExecuTorch][WebGPU] SymInt arithmetic ops (add/sub/mul/floordiv) for dynamic shapes #20573

Make sigmoid and select_copy serve any live shape from one graph; fix select's last-token index under dynamic shapes.

Problem: Both ops baked their dispatch/params/output shape at build() for the max shape. select_copy was worse: a negative index (e.g. -1 for the last token) was normalized against the build-time MAX dim, so at a smaller live S it selected a stale/zero position past the live data — producing wrong (often zero) output.

Solution:

• sigmoid (generic add_unary_op): a resize hook recomputes num_elements/dispatch and sets the output cur_dims (shape-preserving).
• select_copy: KEEP the raw (possibly negative) index at build; a resize hook re-resolves it against the LIVE dim, recomputes the output dims (= input minus dim), rebuilds the out/in TensorMeta UBOs and the dispatch.
• Both keep their uniform buffer(s) alive via own_uniform_buffer.

Implementation:

• The select out/in meta is rebuilt from synthetic WebGPUTensor{dims} via fill_tensor_meta (reads only .dims).
• Mirrors Vulkan per-op resize_*_node.

Constraints: Behavior-neutral on static graphs (hooks fire only when an input's live shape differs from the max). No kernel/WGSL/numerics change.

Co-authored-with: Claude Code.

Differential Revision: D109906095

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20578

• 📄 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, 1 Unrelated Failure

As of commit 3d8d933 with merge base 55a71e6 ():

NEW FAILURE - The following job has failed:

• pull / test-static-llama-qnn-linux (stories_260k_bc) / linux-job (gh)
  test_llama_stories_260k

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• pull / test-qnn-testsuite-linux / test-backend-linux (qnn, models) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

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.