NVE energy conservation in neighbor rebuilds

[laserkelvin]

ALCHEMI Toolkit Pull Request

Description

This PR closes #151, where NVE simulations with wrapped periodic boundaries eventually invalidate neighbor list caches once atoms are wrapped, but continue to compute the energy with the wrong set of periodic shifts.

The fix changes the rebuild check to use raw cartesian displacements instead

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Performance improvement
• Documentation update
• Refactoring (no functional changes)
• CI/CD or infrastructure change

Related Issues

Changes Made

• Add a slow unit test that checks for NVE divergence with LJ after 1000 steps; this is less than the MWE in 🐛[BUG]: NVE energy drift with NeighborListHook(skin>0) + WrapPeriodicHook: position folds bypass the Verlet-skin rebuild check #151, but should be sufficient to show some kind of drift at that point
• Removes the compat import, and just uses batch_neighbor_list_rebuild_inplace from nvalchemiops

Testing

• Unit tests pass locally (make pytest)
• Linting passes (make lint)
• New tests added for new functionality meets coverage expectations?

Checklist

• I have read and understand the Contributing Guidelines
• I have updated the CHANGELOG.md
• I have performed a self-review of my code
• I have added docstrings to new functions/classes
• I have updated the documentation (if applicable)

Additional Notes

Tip

This repository uses Greptile, an AI code review service, to help conduct
pull request reviews. We encourage contributors to read and consider suggestions
made by Greptile, but note that human maintainers will provide the necessary
reviews for merging: Greptile's comments are not a qualitative judgement
of your code, nor is it an indication that the PR will be accepted/rejected.
We encourage the use of emoji reactions to Greptile comments, depending on
their usefulness and accuracy.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[laserkelvin]

/ok to test 83b991e

[greptile-apps]

Greptile Summary

This PR switches neighbor-list rebuild detection to raw Cartesian displacement so periodic wrapping invalidates cached shifts.

• Removes the local in-place rebuild custom operation and its distributed passthrough registration.
• Uses the upstream nvalchemiops rebuild detector directly.
• Adds periodic-fold and long-running NVE energy-conservation regression coverage.

Important Files Changed

Filename	Overview
nvalchemi/hooks/neighbor_list.py	Replaces minimum-image rebuild detection with upstream raw-Cartesian displacement checks so wrapped coordinates refresh cached periodic shifts.
nvalchemi/dynamics/_ops/neighbor_list_rebuild.py	Removes the superseded local in-place neighbor-list rebuild custom operation.
nvalchemi/distributed/shard_wrappers.py	Removes passthrough registration for the deleted custom rebuild operation.
test/hooks/test_neighbor_list_hook.py	Adds focused coverage proving that periodic folding triggers a rebuild and produces a valid cached shift.
test/hooks/test_neighbor_list_nve_conservation.py	Adds a deterministic slow regression test comparing NVE energy drift with skin caching against full rebuilds.

_{Reviews (1): Last reviewed commit: "refactor(hooks): simplify rebuild import" | Re-trigger Greptile}