parakeet portable mode works on cuda server

[Gasoonjia]

Previously we hardcoded a wrong number of mel for model export, leading to tensor resize issue when we running parakeet portable mode on cuda server.

This PR makes the size of tensor used for model export decided by user-specific max length of audio.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/16674

• 📄 Preview Python docs built from this PR

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

❌ 3 New Failures, 2 Cancelled Jobs, 1 Unrelated Failure

As of commit 96d1e08 with merge base 3341474 ():

NEW FAILURES - The following jobs have failed:

• pull / test-samsung-models-linux / linux-job (gh)
  test_inception_v3_fp16
• pull / unittest / macos / macos-job (gh)
  examples/models/test/test_export.py::ExportTest::test_mv3_export_to_executorch
• Test CUDA Builds / test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t dd0fddd1d628583be19c90ae228c79be095ffc436e18f619efee57df3171e3fd /exec failed with exit code 1

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

• pull / test-static-llama-qnn-linux (stories_110m) / linux-job (gh)
  ##[error]The operation was canceled.
• pull / test-static-llama-qnn-linux (stories_260k_bc) / linux-job (gh)
  ##[error]The operation was canceled.

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

• Test Metal Backend / test-model-metal-e2e (openai, whisper-large-v3-turbo, non-quantized) / macos-job (gh) (matched macos rule in flaky-rules.json)
  File doesn't exist

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.

[JacobSzwejbka]

It seems like this PR just makes the max audio length user settable. Its not clear to me why this specifically was impacting your ability to run with cuda acceleration.

[mergennachin]

Keep it 600 seconds like before

[Gasoonjia]

It seems like this PR just makes the max audio length user settable. Its not clear to me why this specifically was impacting your ability to run with cuda acceleration.

The thing is we miscalculated the num of mel we need to use for export from 1000 to 100; therefore when I working on portable mode on cuda server it can not support it due to the input is too large for the pte
not super sure why it can work on Mac (Claude said cpu PyTorch is not that strict so it can work but gpu PyTorch is stricter)
This pr makes the export file use the correct tensor for export, and make sure the miscalculation will not happen again