[FEAT] Perf Profiler Update

[jeromeku]

@msaroufim

Same as #495 except rebased on main.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/690

• 📄 Preview Python docs built from this PR

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

✅ No Failures

As of commit b39443a with merge base 7f0621d ():
💚 Looks good so far! There are no failures yet. 💚

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

[jerryzh168]

@jeromeku really happy to see you updating this again, we were just discussing about this today. I think this is very important work and hope to merge this soon, I think one major thing before merging here is just to integrate the PERF_COUNTER etc. in torchao/_models/llama/perf_profile.py directly with https://github.com/pytorch/ao/blob/main/torchao/_models/llama/generate.py so we don't have duplications

cc @msaroufim @HDCharles

[msaroufim]

Let's do code unification in a later PR

[msaroufim]

Thanks!!