NXP Backend: Add pass to remove unnecessary Quantize/Dequantize nodes.

[MartinPavella]

Summary

This PR adds an edge dialect pre-processing pass to remove some Q/DQ nodes. This enables some non-delegated nodes (which run on the CPU) to run in directly in int8 and avoid the QDQ compute overhead. This improves the inference speed (by eliminating the need to artificially quantize and de-quantize input and output values.

Test plan

Unit tests provided.

cc @robert-kalmar

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15148

• 📄 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! (1 Unrelated Failure)

As of commit b2831c3 with merge base 3b1aeda ():

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• pull / unittest-arm-backend-with-no-fvp (test_pytest_models) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.

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

[MartinPavella]

@pytorchbot label "module: nxp" "release notes: nxp"

[robert-kalmar]

Update the Summary, the pass has different intention:

This PR adds an edge dialect pre-processing pass to remove some Q/DQ nodes. This enables some non-delegated nodes (which run on the CPU) to run in directly in int8 and avoid the QDQ compute overhead. This improves the inference speed (by eliminating the need to artificially quantize and de-quantize input and output values.