Enable buck-native x86 simulator test for QNN op tests (#20494)

[billmguo]

Summary:

The QNN operator tests in backends/qualcomm/tests/test_qnn_delegate.py already support host x86_64 simulator execution via --enable_x86_64, but were only runnable as a standalone argparse script after a CMake build. There was no buck test path: test_qnn_delegate is a python_library (not a test target), and the host qnn_executor_runner binary was Android-only.

This change wires up a buck-native, internal-only test target //executorch/backends/qualcomm/tests:test_qnn_delegate_x86 that runs the FP16 and quantized operator suites (TestQNNFloatingPointOperator, TestQNNQuantizedOperator) on the x86 QNN simulator with no device and no CMake build tree.

Changes:

• backends/qualcomm/runtime/targets.bzl: add CXX to the pal library platforms (gated if is_fbcode()). It already ships pal/src/linux/*.cpp, but was gated to [ANDROID], which blocked any host build of :runtime (an exported dep).
• examples/qualcomm/executor_runner/targets.bzl: add CXX to qnn_executor_runner (gated if is_fbcode()) so the host runner binary builds, and add //executorch/kernels/portable:generated_lib to its deps. The CMake runner links full_portable_ops_lib + quantized_ops_lib; the Buck runner had only the quantized lib, so ops that leave a CPU-fallback node (e.g. acos -> aten::asin.out, cast, index_copy, index_put, logical_and, avg_pool1d) aborted the runner with a "Missing operator" error. The CXX (host) surface is gated to is_fbcode() because it is only used by the internal x86 simulator test; in OSS, CXX includes macOS (no QNN host libs), so the host runner/pal stay Android-only there, restoring the original OSS build surface.
• backends/qualcomm/tests/test_qnn_delegate_x86.py (new): under buck test the file's argparse __main__/setup_environment() never runs, so this wrapper sets the equivalent TestQNN class attributes (enable_x86_64, backend, soc_model) at import and subclasses the operator TestCases so the runner discovers them.
• backends/qualcomm/tests/BUCK: add the test_qnn_delegate_x86 python_test, gated behind runtime.is_oss via a top-level conditional expression (the BUCK dialect forbids top-level if/def). It stays out of the OSS graph to preserve the test-qnn-buck-build-linux CI signal. The QNN x86 SDK libs and host runner are supplied via env (QNN_SDK_ROOT, LD_LIBRARY_PATH, QNN_EXECUTOR_RUNNER). CUDA_VISIBLE_DEVICES is forced empty because these tests are CPU-only (calibration, AOT compile, x86 sim); without it, parallel test shards each grabbed CUDA and exhausted GPU memory.

The fbcode and xplat copies of all four files are kept byte-identical per the existing twin convention.

Differential Revision: D109606746

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20494

• 📄 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:

• [ROCm] MI350 CI runner label rename: rebase PRs using old linux.rocm.gpu.gfx950.* labels
• [Housekeeping] Rename uw1 cluster to meta-prod-aws-uw1

❌ 3 New Failures, 1 Unrelated Failure

As of commit 3545844 with merge base 45a14b9 ():

NEW FAILURES - The following jobs have failed:

• pull / test-qnn-buck-build-linux / linux-job (gh)
  RuntimeError: Command docker exec -t 34619dce3c6c10238744672bddea1f24fd32b7df13f1df75a808ce85b3ad5d3c /exec failed with exit code 3
• pull / unittest-buck / linux / linux-job (gh)
  RuntimeError: Command docker exec -t 1c18cf0eee09ab48372b41554e92e55318f2c88dc4df3a3557a7a6db4b76bf1e /exec failed with exit code 3
• pull / unittest-buck / macos / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 3

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-llama-runner-qnn-linux (fp32, qnn_16a16w, qnn) / linux-job (gh) (trunk failure)

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

[linux-foundation-easycla]

• ❌ - login: @billmguo / name: Min Guo. The commit (3545844) is not authorized under a signed CLA. Please click here to be authorized. For further assistance with EasyCLA, please visit our EasyCLA portal and chat with our support bot.

[meta-codesync]

@billmguo has exported this pull request. If you are a Meta employee, you can view the originating Diff in D109606746.

[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.