NXP backend: Replace pass to fuse activations functions with joint quantization with activation

[roman-janik-nxp]

Summary

This PR replaces optimizations 'fuse_activation_functions.py' by quantization of Conv 2D and Linear ops together with fusable activations - selected activations supported by Neutron (Relu, Relu6, Sigmoid, Tanh). Logic is determined by target specs, now supporting Neutron-C. Tests updated. Relu has now non-shared, standalone quantization.

Test plan

Unit tests provided (test_edge_passes.py, test_quantizer.py).

cc @robert-kalmar @JakeStevens @digantdesai

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14816

• 📄 Preview Python docs built from this PR

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

❌ 4 New Failures

As of commit 8d9107b with merge base 2c603e4 ():

NEW FAILURES - The following jobs have failed:

• pull / test-samsung-models-linux / linux-job (gh)
  RuntimeError: Command docker exec -t a78b19dc8dbe5bb37605691e67c3138a2560ca2b44997eb1755a812c990d5d98 /exec failed with exit code 1
• pull / unittest-editable / macos / macos-job (gh)
  backends/xnnpack/test/ops/test_conv1d.py::TestConv1d::test_qs8_conv1d_batchnorm_seq
• pull / unittest-nxp-neutron / linux-job (gh)
  RuntimeError: Command docker exec -t a28a911acf64b2716ce18362d6c01d779b21d8de1b5855e1ba7c9ccd9af459dd /exec failed with exit code 1
• Test CUDA Builds / test-voxtral-cuda-e2e / linux-job (gh)
  RuntimeError: Command docker exec -t 574b6952338209da15edb19333cc684ba1964b26189f874c3421a7c0847e87ad /exec failed with exit code 2

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

[roman-janik-nxp]

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

[MartinPavella]

Is this necessary?
I would expect that since the Move*AuxiliaryOperatorIntoSeparateQDQClusterPass now supports these operators, they should be the main nodes within their own QDQ clusters.

[roman-janik-nxp]

I've looked into into it. The reason is it's in the QDQClusterRecognizer AUXILIARY_OPS list not only for the use in partitioner, but because the QDQClusterRecognizer is also used in the Move*AuxiliaryOperatorIntoSeparateQDQClusterPass:

executorch/backends/nxp/edge_passes/move_auxiliary_operator_into_separate_qdq_cluster_pass.py

Line 233 in 33f153a

cluster = QDQClusterRecognizer().get_qdq_cluster(main_cluster_node)