forcefields: allow specifying custom calculator in forcefield makers

[lan496]

Summary

• Enable specifying a dictionary recognizable by atomate2.forcefields.utils.ase_calculator at ForceFieldMixin.force_fields_name parameter. This will be useful to integrate atomate2.forcefields with recently released MLIPs or in-house developed MLIPs.
• Add document on how to specify forcefields in docs/user/codes/forcefields.md

Backward compatibility

The following table summarizes the stored attributes in Maker and schemas for forcefields.
For str or MLFF input, the behavior does not change.

For a dict input,

• It is regarded as the placeholder MLFF to keep the other behavior
• ForceFieldMixin.calculator_meta parameter (initialized in __post_init__) is newly introduced to remember the dict information
• ForceFieldMixin.ase_calculator_name property is newly introduced to keep schemas for ForceFieldMeta.forcefield_name. Letting calc_cls be a decoded calculator function from the dict, calc_cls.__name__ is stored
• The package version is inferred from calc_cls.__module__

Input for ForceFieldMixin.force_field_name	'CHGNet' | 'MLFF.CHGNet' | MLFF.CHGNet	{"@module": "chgnet.model.dynamics", "@callable": "CHGNetCalculator"}
ForceFieldMixin.force_field_name after __post__init__	'MLFF.CHGNet'	'MLFF.Forcefield'
ForceFieldMixin.mlff (introduced in #1360)	MLFF.CHGNet	MLFF.Forcefield
ForceFieldMixin.calculator_meta (added in this PR)	MLFF.CHGNet	{"@module": "chgnet.model.dynamics", "@callable": "CHGNetCalculator"}
ForceFieldMixin.ase_calculator_name (added in this PR) / ForceFieldMeta.forcefield_name	'MLFF.CHGNet'	'CHGNetCalculator'
ForceFieldMeta.forcefield_version	version('chgnet')	version('chgnet')

Checklist

Work-in-progress pull requests are encouraged, but please put [WIP] in the pull request
title.

Before a pull request can be merged, the following items must be checked:

• Code is in the standard Python style.
  The easiest way to handle this is to run the following in the correct sequence on
  your local machine. Start with running ruff and ruff format on your new code. This will
  automatically reformat your code to PEP8 conventions and fix many linting issues.
• Doc strings have been added in the Numpy docstring format.
  Run ruff on your code.
• Type annotations are highly encouraged. Run mypy to
  type check your code.
• Tests have been added for any new functionality or bug fixes.
• All linting and tests pass.

Note that the CI system will run all the above checks. But it will be much more
efficient if you already fix most errors prior to submitting the PR. It is highly
recommended that you use the pre-commit hook provided in the repository. Simply run
pre-commit install and a check will be run prior to allowing commits.

[lan496]

This is not an essential fix to the PR itself

[lan496]

I have confirmed test-force-field-notebook passes, but I have no idea why tests/abinit/jobs/test_base.py fails with this unrelated PR 🤷‍♂️
https://github.com/materialsproject/atomate2/actions/runs/20565430185/job/59062899029?pr=1362

[JaGeo]

@lan496 Yeah, it has nothing to do with your PR. I have restarted the tests manually.

[JaGeo]

@lan496 thanks!

One more question: what happens if the phonon maker is not initalized via the from_... method but rather manually via _init_? Would the code still work without issue?

[lan496]

For PhononMaker, ASE calculator information is taken from PhononMaker.phonon_displacement_maker: ForceFieldStaticMaker (this attribute is not optional). So, if PhononMaker is properly constructed, it should work as expected.

[JaGeo]

@esoteric-ephemera I will merge. It is backwards compatible and should work well with all previous examples. Feel free to comment more after the holidays in case I overlooked anything.

[JaGeo]

@lan496 Thank you! I will merge!

[lan496]

@JaGeo Thank you so much!