XNNPACK: Kleidi QP8 and SME2

[digantdesai]

Summary

• enable qp8 dqlinear paths at runtime
• enable xnnpack arm sme2 isa build flag (XNNPACK_ENABLE_ARM_SME is already ON)

Benchmark and SME2 data

• Not so rigorous benchmarking on an M4 using
  executor_runner --model_path vit_xnnpack_q8.pte --num_executions 100
  with and without SME2 flag shows, and w/ SME2 it is consistently faster.
• Verified kleidi SME2 kernel is used through Instrumentation on M4
• LLM packing perf is still needs to be validated with QP8.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/13887

• 📄 Preview Python docs built from this PR

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

❌ 2 New Failures, 3 Unrelated Failures

As of commit b0942ee with merge base c393d17 ():

NEW FAILURES - The following jobs have failed:

• Build documentation / build (buck2) / Build doc (gh)
• trunk / test-huggingface-transformers-coreml (qwen3-1.7b|xnnpack|--quantize) / macos-job (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 138

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

• pull / test-vulkan-operators-linux / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
• trunk / test-arm-cortex-m-size-test (zephyr-preset) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
• trunk / test-qnn-model (fp32, mv2) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

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

[mergennachin]

Godspeed!