mps + compile + quant: introduce mps_op_lib

[manuelcandales]

Needed to enable the use of torchao quantized lowbit kernels with mps compile

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchchat/1530

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• [Infra] Jobs got intermittently cancelled/fail midway checkout

❌ 4 New Failures, 2 Cancelled Jobs, 8 Unrelated Failures

As of commit 738d400 with merge base f7c4f14 ():

NEW FAILURES - The following jobs have failed:

• pull / runner-et (macos-14-xlarge) (gh)
  ImportError: cannot import name 'EmbeddingQuantizer' from 'torchao.experimental.quant_api' (/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/torchao/experimental/quant_api.py)
• pull / test-torchao-experimental-cpp (macos-14-xlarge) (gh)
  RuntimeError: dictionary changed size during iteration
• pull / test-torchao-experimental-et (macos-14-xlarge) (gh)
  RuntimeError: dictionary changed size during iteration
• pull / test-torchao-experimental-python (macos-14-xlarge) (gh)
  RuntimeError: dictionary changed size during iteration

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

• pull / runner-et (16-core-ubuntu) (gh)
  ##[error]The operation was canceled.
• pull / test-gpu-compile (cuda, stories15M) / linux-job (gh)
  ##[error]The operation was canceled.

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

• pull / test-gpu-aoti-bfloat16 (cuda, stories15M) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• pull / test-gpu-aoti-float16 (cuda, stories15M) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• pull / test-gpu-aoti-float32 (cuda, stories15M) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• pull / test-gpu-eval-sanity-check (cuda, stories15M) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• Run parallel prefill / test-cuda / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• Run parallel prefill / test-sdpa-backends-export / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• Run the aoti runner with CUDA using stories / test-runner-aot-cuda / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• Run the README instructions - with stories / test-multimodal-cpu / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.

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