Metal backend: Add AOTI shims for memory management

[manuelcandales]

Implement the following shims:

• aoti_torch_create_tensor_from_blob_v2
• aoti_torch_empty_strided
• aoti_torch_delete_tensor_object
• aoti_torch_copy_
• aoti_torch__reinterpret_tensor

[manuelcandales]

Stack from ghstack (oldest at bottom):

• Add Metal backend build system and runtime integration #15024
• Metal backend: Add operator implementations #15023
• Metal backend: Implement the AOTI MPS shim #15022
• -> Metal backend: Add AOTI shims for memory management #15021
• Add Metal backend core ETMetal runtime. #15020
• Add Metal backend type definitions and utilities #15019
• Add Metal backend Python preprocessing, partitioning, and tests #15015
• [ET][Metal] Update aoti_common with additional AOTI functions needed by Metal backend #15003

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15021

• 📄 Preview Python docs built from this PR

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

❗ 2 Active SEVs

There are 2 currently active SEVs. If your PR is affected, please view them below:

• Macos CI runners unavailable
• [ROCm][CI] Machines under the label linux.rocm.gpu.2 are undergoing maintenance.

❌ 5 New Failures, 1 Cancelled Job

As of commit f12117b with merge base 896178e ():

NEW FAILURES - The following jobs have failed:

• pull / test-qnn-wheel-packages-linux (3.10) / linux-job (gh)
  RuntimeError: Command docker exec -t 75b59d8cb1c39ebd5065beebde4076e796352380931fe186a85c2bd63db2462a /exec failed with exit code 1
• pull / test-qnn-wheel-packages-linux (3.11) / linux-job (gh)
  RuntimeError: Command docker exec -t 054b2ce95fc5e3d765f5c757d3eb1b80f60fe74c9db43079b10118b0344359de /exec failed with exit code 1
• pull / test-qnn-wheel-packages-linux (3.12) / linux-job (gh)
  RuntimeError: Command docker exec -t f26471f294221bcb38c14744f7b7c0f663ca0297eaeaaa54d4c3d754d675b38f /exec failed with exit code 1
• pull / unittest / macos / macos-job (gh)
  export/tests/test_target_recipes.py::TestTargetRecipes::test_linear_model
• pull / unittest-arm-backend-with-no-fvp (test_pytest_ops) / linux-job (gh)
  RuntimeError: Command docker exec -t 4f056b70126d24afbd8607958c55d6df949bd1c08827964f2e81862fd4c73689 /exec failed with exit code 1

CANCELLED JOB - The following job was cancelled. Please retry:

• Build Windows Wheels / generate-matrix / generate (gh)

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

[mergennachin]

Why not use the function?

int32_t mps_device_type = aoti_torch_device_type_mps();  // Returns 13
if (device_type == mps_device_type) {
  ...
}

[mergennachin]

returning early causes memory leak, no?

[mergennachin]

Maybe explicitly implement metal_free_buffer to properly clean up metal buffers and call the method whenever appropriate

[mergennachin]

duplicate declaration