Qualcomm AI Engine Direct - Adding QNN backend support for the sort core ATen op

[qti-horodnic]

Summary

Added support for the aten.sort core ATen op for QNN HTP backend.
Implements aten.sort.default delegation to QNN HTP by reusing the existing TopK op with k = dim_size. Supports both ascending and descending sort order.

Note that this op faces the same limitations as the existing TopK op:

• Only supports sorting along the last dimension (QNN constraint).
• Sort indices are output as int32; consuming them with gather in quantized models may cause dtype mismatches. Added a floating point sort_and_index test for this use-case.

Test plan

pytest backends/qualcomm/tests/rework/htp/op/v68/test.py -k "test_sort"


python backends/qualcomm/tests/test_qnn_delegate.py TestQNNQuantizedOperator.test_qnn_backend_sort --soc_model SM8750 --host aisw-vm15-labsd --device 545ee4aa --build_folder build-android

python backends/qualcomm/tests/test_qnn_delegate.py TestQNNFloatingPointOperator.test_qnn_backend_sort --soc_model SM8750 --host aisw-vm15-labsd --device 545ee4aa --build_folder build-android

python backends/qualcomm/tests/test_qnn_delegate.py TestQNNFloatingPointOperator.test_qnn_backend_sort_and_index --soc_model SM8750 --host aisw-vm15-labsd --device 545ee4aa --build_folder build-android

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21386

• 📄 Preview Python docs built from this PR

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

⚠️ 12 Awaiting Approval

As of commit b07b564 with merge base ff0f92e ():

AWAITING APPROVAL - The following workflows need approval before CI can run:

• Build documentation (gh)
• Build Presets (gh)
• Lint (gh)
• periodic (gh)
• pull (gh)
• Test ARM Backend (gh)
• Test CoreML Backend (gh)
• Test QNN Backend (gh)
• Test QNN Windows MSVC build (gh)
• Test Vulkan Backend (gh)
• Test WebGPU Backend (gh)
• Test XNNPACK Backend (gh)

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

[qti-horodnic]

@pytorchbot label "release notes: qualcomm"