Add unit tests to consolidate selective build APIs

[TejasweeSulekh]

Summary

This PR officially resolves the selective build API consolidation requested in the original issue.

The CMake and Python logic (gen_oplist.py) already support parsing --root_ops and --model_file_path simultaneously, naturally merging the resulting operator sets. However, this behavior was previously untested, leaving it vulnerable to regression. This PR adds the missing unit test test_gen_op_list_with_model_and_root_ops to officially lock in and prove this behavior.

Fixes #11921

Testing Plan

Manually verified that the new unit test correctly mocks the operator lists and asserts the merged set without throwing a mutual exclusivity error.

# Activated virtual environment
source .venv/bin/activate

# Ran the codegen test suite to verify the new test passes
pytest codegen/tools/test/test_gen_oplist.py

cc @larryliu0820 @GregoryComer @nil-is-all

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/23176

• 📄 Preview Python docs built from this PR

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

⏳ 1 Pending, 1 Unrelated Failure

As of commit 7ba7fd5 with merge base 96c621d ():

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

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

• pull / test-arm-backend-no-driver (test_pytest_models_tosa) / linux-job (gh) (trunk failure)
  RuntimeError: Command bash /__w/_temp/exec_script failed with exit code 1

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

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[metascroy]

cc @JacobSzwejbka for core runtime

[TejasweeSulekh]

Hey team,

It looks like the CI is failing across the board during the initial environment setup phase.

Specifically, install_requirements.py is failing to fetch torchao==0.18.0.dev20260729 (No matching distribution found).

My unit tests haven't been executed yet. Let me know once the dependencies are fixed on main and I will gladly rebase this branch

[nil-is-all]

Hey team,

It looks like the CI is failing across the board during the initial environment setup phase.

Specifically, install_requirements.py is failing to fetch torchao==0.18.0.dev20260729 (No matching distribution found).

My unit tests haven't been executed yet. Let me know once the dependencies are fixed on main and I will gladly rebase this branch

Hi @TejasweeSulekh, we've just fixed the dependency issue. Please do rebase and give me a ping. I'll trigger CI tests again

[TejasweeSulekh]

Hey @nil-is-all, I rebased the repo with the latest commit. From my end the following are working well:

## Basically the test being added here
pytest codegen/tools/test/test_gen_oplist.py 

## Testing the cxx side
bash ./test/run_oss_cpp_tests.sh

But this test is failing for some reason and I can't really be sure that this is because of the changes made in this PR.

## Testing EXIR side of the library
pytest exir/tests/ 

Where the error indicates:

RuntimeError: Missing out variants: {'quantized_decomposed::dequantize_per_tensor', 'quantized_decomposed::add'...}

Should I wait for this error to be fixed or is the current state of the PR acceptable?

[nil-is-all]

Hey @nil-is-all, I rebased the repo with the latest commit. From my end the following are working well:

## Basically the test being added here
pytest codegen/tools/test/test_gen_oplist.py 

## Testing the cxx side
bash ./test/run_oss_cpp_tests.sh

But this test is failing for some reason and I can't really be sure that this is because of the changes made in this PR.

## Testing EXIR side of the library
pytest exir/tests/ 

Where the error indicates:

RuntimeError: Missing out variants: {'quantized_decomposed::dequantize_per_tensor', 'quantized_decomposed::add'...}

Should I wait for this error to be fixed or is the current state of the PR acceptable?

Thanks for the ping. I'll run the tests on my end and keep you posted on the error.

[TejasweeSulekh]

Hey @nil-is-all, sorry, but there was lintrunner formatting issue. I fixed it and updated the PR. Whenever convenient can you please start the CI tests?