Update ExecuTorch pin to latest viable/strict 3/28/2025 (#150308)

[huydhn]

From latest viable/strict: https://hud.pytorch.org/hud/pytorch/executorch/viable%2Fstrict/1?per_page=50

Fixes #144480

This commit has important CI stability fixes, such as pytorch/executorch#9561 and pytorch/executorch#9634

Pull Request resolved: #150308
Approved by: https://github.com/jathu, https://github.com/malfet

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/153750

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

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

❌ 6 New Failures, 5 Cancelled Jobs

As of commit 27ac4a8 with merge base 924a247 ():

NEW FAILURES - The following jobs have failed:

• Build libtorch docker images / build-docker-cpu (gh)
  Final attempt failed. Child_process exited with error code 1
• Build libtorch docker images / build-docker-cuda (12.8) (gh)
  Final attempt failed. Child_process exited with error code 1
• Build libtorch docker images / build-docker-rocm (6.3) (gh)
  Final attempt failed. Child_process exited with error code 1
• Build manywheel docker images / build-docker-cpu-cxx11-abi (gh)
  Final attempt failed. Child_process exited with error code 1
• Build manywheel docker images for s390x / build-docker-cpu-s390x (gh)
  ValueError: SELinux policy is not managed or store cannot be accessed.
• docker-builds / docker-build (linux.12xlarge, pytorch-linux-jammy-py3.12-halide) (gh)
  Final attempt failed. Child_process exited with error code 1

CANCELLED JOBS - The following jobs were cancelled. Please retry:

• Build libtorch docker images / build-docker-cuda (11.8) (gh)
  ##[error]The operation was canceled.
• Build libtorch docker images / build-docker-cuda (12.4) (gh)
  ##[error]The operation was canceled.
• Build libtorch docker images / build-docker-cuda (12.6) (gh)
  Final attempt failed. Child_process exited with error code 1
• Build libtorch docker images / build-docker-rocm (6.2.4) (gh)
  Final attempt failed. Child_process exited with error code 1
• inductor / unit-test / linux-jammy-cpu-py3.12-gcc11-inductor-halide / build (gh)

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

[huydhn]

@atalman I just realize that we would need to cherry pick several changes to make this green as this ExecuTorch commit pin depends on a torchtune nightly that has been removed from the index https://github.com/pytorch/pytorch/actions/runs/15078606465/job/42392442383#step:22:611. However, I want to avoid doing so because this would mean cherry picking too many changes. So, I think it's better to temporarily add torchtune==0.6.0.dev20250310 back to the nightly index instead

[huydhn]

@atalman Here is the index script update pytorch/test-infra#6639, I guess it's not counted as a cherry pick nor a release only change. It's a weird case

[atalman]

lgtm