Metal backend: enable linear with bias

[manuelcandales]

This pull request introduces a new Metal-specific graph transformation pass for the Apple Metal backend, aimed at decomposing aten.linear operations into simpler operations to improve compatibility and avoid reliance on addmm. It also adds the new pass to the backend's custom passes, exposes it in the package, and introduces a corresponding test module.

Metal Backend Linear Decomposition Pass:

• Added a new DecomposeLinearPass in decompose_linear_pass.py, which rewrites aten.linear nodes in the computation graph into a sequence of matmul and add operations. For 2D inputs, it temporarily unsqueezes to 3D to avoid triggering addmm, then squeezes back to 2D, ensuring compatibility with Metal's capabilities.
• Registered the new DecomposeLinearPass in the Metal backend's get_custom_passes method, so it is applied during compilation.
• Exposed DecomposeLinearPass in the passes module's __init__.py for easier imports and future extensibility.

Testing and Module Registry:

• Added a new LinearWithBias module to the module registry in the test suite, providing a model that exercises the new pass and ensuring coverage for linear layers with bias.

[manuelcandales]

Stack from ghstack (oldest at bottom):

• Metal backend: fix bugs in qmv_impl #17188
• Metal backend: import qmv_impl from MLX #17187
• Metal backend: int4 refactor checks and logs #17186
• Metal backend: test 4-bit linear with bias #17127
• Metal backend: enable quantization in test_modules #17118
• Metal backend: support 4-bit linear #17117
• Metal backend: remove parakeet linear decomp #17116
• -> Metal backend: enable linear with bias #17115
• Metal backend: test modules #17076

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/17115

• 📄 Preview Python docs built from this PR

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

❌ 2 New Failures, 11 Cancelled Jobs, 2 Pending, 1 Unrelated Failure

As of commit ed4dcee with merge base ba6de95 ():

NEW FAILURES - The following jobs have failed:

• pull / unittest-buck / linux / linux-job (gh)
  RuntimeError: Command docker exec -t db87a0116fde2cd926bc5c3a3561f94cbca87ff01365de63aef66e4467e094a8 /exec failed with exit code 3
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 69ed1cddafb2c35ff062b92e02d770f164260256ad3eb6271f05f6a122eb9835 /exec failed with exit code 1

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

• Apple / build-benchmark-app / macos-job (gh)
• Apple / build-demo-ios / macos-job (gh)
• Apple / set-version (gh)
• Build Presets / apple (ios-simulator) / build (gh)
• Build Presets / apple (ios) / build (gh)
• Build Presets / apple (llm) / build (gh)
• Build Presets / apple (macos) / build (gh)
• Build Presets / apple (profiling) / build (gh)
• Build Presets / apple (pybind) / build (gh)
• Lint / Get changed files / get-changed-files (gh)
• periodic / gather-models (gh)

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• pull / unittest-buck / macos / macos-job (gh) (trunk failure)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 3

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