[mps] Add offsets to enable aoti

[angelayi]

Stack from ghstack (oldest at bottom):

• [mps] Add cshim for torchao mps ops #2485
• -> [mps] Add offsets to enable aoti #2484

In AOTI we allocate all weights in one blob and index into the blob based on the offsets. This caused an accuracy error when running the lowbit kernels since it assumes all the offsets are 0.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2484

• 📄 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

As of commit 6f8aff7 with merge base a2c5ca1 ():

NEW FAILURES - The following jobs have failed:

• Run Regression Tests / test-nightly (CPU Nightly, linux.4xlarge, --pre torch --index-url https://download.pytorch.org/wh... / linux-job (gh)
  test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_True_dynamic_True
• Run Regression Tests / test-nightly (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch --index-url https://downloa... / linux-job (gh)
  test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_quant_semi_sparse_compile_False
• Run TorchAO Experimental Tests / test-mps-ops (macos-m1-stable) (gh)
  Process completed with exit code 127.

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