[ET-VK][ops] Optimize narrow q8ta convolution workgroups

[SS-JIA]

Stack from ghstack (oldest at bottom):

• [ET-VK][ops] Bound q8ta im2col scratch memory #22536
• -> [ET-VK][ops] Optimize narrow q8ta convolution workgroups #22535

Use narrow local workgroups for regular and depthwise q8ta convolutions whose output width underfills existing workgroups. This reduces inactive invocations in SceneX small-spatial layers. Authored with Codex.

Differential Revision: D118709586

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/22535

• 📄 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 e125c30 with merge base 2c1da32 ():

NEW FAILURE - The following job has failed:

• pull / test-samsung-models-linux / linux-job (gh)
  RuntimeError: Command docker exec -t cca73ddb884fcaed287a7b465569dc18dbdbad14d251413af22d3d34a369fa36 /exec 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.