[ExecuTorch][WebGPU] test: q4gsw shmem-GEMM prefill golden configs

[JulianCloudNTH]

Stack from ghstack (oldest at bottom):

• -> [ExecuTorch][WebGPU] test: q4gsw shmem-GEMM prefill golden configs #20606
• [ExecuTorch][WebGPU] q4gsw prefill: shared-memory tiled GEMM, shape-routed (+150–303% large-K/N) #20605

Add native fp64-golden configs that exercise the shared-memory tiled GEMM prefill route.

The existing linear_q4gsw prefill configs (q_proj_4k/kv_proj_4k, K=2048, N<=2048) all route to the register-tiled GEMM, so the new shmem GEMM route (K >= 4096 || N >= 4096) had no coverage. This adds three configs — kept in lockstep between test_quantized_linear.py CONFIGS and the C++ kQ4gswConfigs table — that exercise it (the op lands in the stacked diff below):

• gate_proj_pf (M128 K2048 N8192) — shmem via large N (gate/up prefill).
• down_proj_pf (M128 K8192 N2048) — shmem via large K (down prefill), looser big-K tol.
• shmem_edge (M130 K4096 N2056) — partial 32-tile bounds (M and N not multiples of 32).

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

Differential Revision: D110095129

Differential Revision: D110095129

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20606

• 📄 Preview Python docs built from this PR

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

✅ You can merge normally! (2 Unrelated Failures)

As of commit 1164e70 with merge base b3e11d3 ():

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

• pull / test-qnn-testsuite-linux / test-backend-linux (qnn, models) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• Test ARM Backend / test-arm / test-backend-linux (arm_tosa_fp, models) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The process '/usr/bin/git' failed with exit code 1

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.