Add support for splitting in_features in linear layers

[metascroy]

This PR:

• Moves SplitLinearModule to a separate utils.py and extends it to support splitting in_features. I did not find splitting in_features worked very well, at least in my initial experiments.
• Updates readme with latest setting recommendations for Llama1B

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/8715

• 📄 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 b8b9d44 with merge base dfe11f0 ():

NEW FAILURE - The following job has failed:

• Check Labels / Check labels (gh)
  RuntimeError: Error checking labels: PR does not have required labels

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

[github-actions]

This PR needs a release notes: label

If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[metascroy]

@cccclai can I get a review on this?

[cccclai]

I might miss some context - I thought you said splitting linear speeds up the perf in ANE?

[cccclai]

Oh, you said in features, in the description. What did you split to get better perf?

[metascroy]

Yes, so currently landed we have ability to split up out_features in linear modules and this speeds up perf by out 30%.

This diff adds the ability to split in_features in linear modules. It did not speed up perf in the few parameters I tried, but it's nice to have this ability it landed so people can experiment further.

[cccclai]

Looks good, thank you!