Arm backend: Avoid not decomposing linears we reject

[Erik-Lundell]

If a linear is not quantized properly, we will reject it when partitioning. However, if we tell Executorch to not not decompose an op, we are required to partition it. We thus need to figure out if we will partition the linear or not in the ops_not_to_decompose filter function.

Also turn off grad in the arm tester to solve an error that popped up in the GRU model. Since we only do inference, grad is never relevant.

cc @freddan80 @per @zingo @oscarandersson8218 @digantdesai

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15406

• 📄 Preview Python docs built from this PR

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

❌ 2 New Failures, 1 Cancelled Job

As of commit 4ad83fa with merge base c1c5a71 ():

NEW FAILURES - The following jobs have failed:

• pull / test-samsung-models-linux / linux-job (gh)
  RuntimeError: Command docker exec -t 0a0e530a59590b31b851f59af3cdf1b64595f277b2c5217cd9ccc7f5b71d1ce7 /exec failed with exit code 1
• trunk / test-llama-runner-mac (fp32, coreml) / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1

CANCELLED JOB - The following job was cancelled. Please retry:

• trunk / test-models-macos-cpu (mv3, portable) / macos-job (gh)
  ##[error]The operation was canceled.

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