Cortex_m backend: Add permute op

[AdrianLundell]

Since the transpose op doesn't require qparams but still expects input to be int8, the check in quantized_op_fusion_pass is moved from the call_operator level to the _get_replacement level. This way different ops can have different checks.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15848

• 📄 Preview Python docs built from this PR

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

✅ You can merge normally! (2 Unrelated Failures)

As of commit 8487927 with merge base 0769abc ():

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

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

• pull / unittest / linux / linux-job (gh) (trunk failure)
  exir/backend/test/test_lowered_backend_module.py::TestBackendAPI::test_emit_nested_lowered_backend_module
• pull / unittest-editable / linux / linux-job (gh) (trunk failure)
  exir/backend/test/test_lowered_backend_module.py::TestBackendAPI::test_emit_nested_lowered_backend_module

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

[Copilot]

Pull Request Overview

This PR adds permute/transpose operation support to the Cortex_M backend and refactors the quantization parameter check logic in the quantized operation fusion pass. The key change moves the qparams validation from the call_operator level to individual _get_replacement methods, allowing different operations to have different requirements—specifically enabling the transpose operation to work with int8 tensors without requiring quantization parameters.

• Adds new transpose operation for the Cortex_M backend with CMSIS-NN integration
• Refactors qparams checking to be per-operation in quantized_op_fusion_pass.py
• Adds comprehensive test coverage for permute, transpose, and t() operations

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
backends/cortex_m/test/ops/test_transpose.py	New test file with comprehensive test cases for permute, transpose, and t() operations
backends/cortex_m/passes/quantized_op_fusion_pass.py	Moves qparams check from call_operator to individual replacement methods; adds _get_permute_replacement method
backends/cortex_m/ops/operators.yaml	Adds transpose.out operator declaration to the operator schema
backends/cortex_m/ops/operators.py	Implements transpose operation definition, meta function, and functional implementation
backends/cortex_m/ops/op_transpose.cpp	C++ kernel implementation using CMSIS-NN's arm_transpose_s8 function
backends/cortex_m/CMakeLists.txt	Adds op_transpose.cpp to build sources

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.