NXP backend: added aten.split support

[novak-vaclav]

Summary

adds support for aten.split operator

Test plan

tests can be manually run using pytest -c /dev/null backends/nxp/tests/

cc @robert-kalmar @MartinPavella

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/16276

• 📄 Preview Python docs built from this PR

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

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

[novak-vaclav]

@pytorchbot label "release notes: nxp"

[novak-vaclav]

@pytorchbot label "module: nxp"

[Copilot]

Pull request overview

This PR adds support for the aten.split operator in the NXP backend by introducing a new pass that decomposes split operations into slice operations. This enables the backend to handle tensor splitting operations that were previously unsupported.

• Implements DecomposeSplitToSlicesPass to convert split operations into equivalent slice operations
• Adds comprehensive test coverage for different split scenarios including split with size, split with sections, GRU-based splits, and single-chunk edge cases
• Integrates the new pass into the default NXP backend pass pipeline

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

File	Description
backends/nxp/aten_passes/decompose_split_to_slices_pass.py	Implements the core pass logic to decompose split operations into slice operations
backends/nxp/aten_passes/neutron_aten_pass_manager.py	Integrates the new decompose split pass into the default pass pipeline
backends/nxp/tests/models.py	Adds test model classes (GRUModel, SplitWithSize, SplitWithSections) to support split operation testing
backends/nxp/tests/test_decompose_split_to_slices.py	Provides comprehensive test coverage for the split decomposition functionality

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[novak-vaclav]

Fixed the issues found by @MartinPavella, please review them. Thank you!

[novak-vaclav]

Issues fixed. Thanks for the review and please re-review.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 6 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The test ID has a trailing period which is inconsistent with other test IDs in the same parametrization. The ID should be "2D" instead of "2D." for consistency.

Suggested change

	pytest.param((8,), 3, 0, id="1D."),
	pytest.param((4, 8), 5, 1, id="2D."),
	pytest.param((8,), 3, 0, id="1D"),
	pytest.param((4, 8), 5, 1, id="2D"),

[Copilot]

The test ID has a trailing period which is inconsistent with other test IDs in the same parametrization. The ID should be "1D" instead of "1D." for consistency.

[Copilot]

The test ID has a trailing period which is inconsistent with other test IDs in the parametrization. The ID should be "1D" instead of "1D." for consistency.

[Copilot]

The test ID has a trailing period which is inconsistent with other test IDs in the parametrization. The ID should be "2D" instead of "2D." for consistency.

[Copilot]

The error message "Faulty split chunks" is unclear and unhelpful for debugging. Consider providing more context about what went wrong, such as "Expected split chunks to be a list or tuple, but got {type(split_nodes_chunks)}".

Suggested change

	raise RuntimeError("Faulty split chunks")
	raise RuntimeError(
	f"Expected split chunks to be a list or tuple, but got "
	f"{type(split_nodes_chunks).__name__}: {split_nodes_chunks!r}"
	)

[Copilot]

This line assumes that the split node has exactly one user when there's only one chunk, but this assumption isn't validated. If the split node has zero or multiple users, accessing list(split_node.users)[0] could raise an IndexError or retrieve the wrong node. Consider adding validation or using a safer access pattern.

Suggested change

	getitem_node = list(split_node.users)[0]
	split_users = list(split_node.users)
	# Only apply this optimization when the split node has exactly one user.
	# If there are zero or multiple users, we cannot safely assume which one
	# should be rewired to the input, so we skip this transformation.
	if len(split_users) != 1:
	continue
	getitem_node = split_users[0]

[roman-janik-nxp]

@novak-vaclav Please do not use List and Tuple from typing, use list and tuple from base Python. This is possible since Python 3.10 and we use it in out codebase.