Add FHI-aims DFT calculator

[tpurcell90]

Summary

Create an FHI-aims interface into atomate2 (moved from: https://github.com/tpurcell90/atomate2-fhi-aims). This was originally started as a project to incorporate FHI-aims into jobflow as a part of the NOMAD-CoE, working with @davidwaroquiers and @ansobolev. Originally we did not know if this could be incorporated into atomate2, which is why we had a separate repository, but everyone now agrees that we should move it to atomate2.

• FHI-aims interface for base jobs, input generators and schemas
• ASE interface that is aims specific (can be moved into a general framework if everyone agrees)
• Generalization of the phonon workflow originally made for VASP

Additional dependencies introduced (if any)

• ASE: This is needed because currently the only way to interact with FHI-aims in python is through ASE. We are in the process of discussing a new FHI-aims controlled python interface, but I have not gotten any type of general support to start to write that. Additionally the SocketIO calculators are useful for the phonon calculations as it can speed up the SCF convergence in FHI-aims by a factor of ~2. The largest problem with this however, is the delay in getting ASE 3.23 out, which is needed for this code interface
• seekpath: This is used in the common phonon workflows, but I did not see it in the dependencies

Note I did not add these to the pyproject file before discussing

TODO (if any)

• Discussion of ASE dependency
• @JaGeo can you check that my modifications to the phonon workflow did not break the VASP workflow? I don't have a VASP license to check this
• run ruff
• run mypy

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 black on your new code. This will
  automatically reformat your code to PEP8 conventions and removes most issues. Then run
  ruff.
• Docstrings 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.

[JaGeo]

@tpurcell90 @utf I am happy to help. I left one immediate comment there already. For the rest, I need more time.

[utf]

Thanks for the PR. In general this looks really high quality and hopefully there won't be much of a barrier to getting it merged. From a quick scan through, it looks like merging the phonon workflows will be the main challenge. If you're happy to coordinate that aspect @JaGeo that would be great.

In terms of getting the linting to pass, I'd recommend installing pre-commit, with:

pip install pre-commit
pre-commit install  # run inside atomate2 folder

This will automatically run the linting every time you do a commit. To run on the existing files, you can do:

pre-commit run --all

[JaGeo]

• seekpath: This is used in the common phonon workflows, but I did not see it in the dependencies

We have additional dependencies for the phonon runs. It is part of it.

[JaGeo]

@tpurcell90 I was too curious to not look at it.
I left several comments. One major comment that I have: The phonon forcefield workflow should probably also inherit from this new base class if possible: https://github.com/materialsproject/atomate2/blob/main/src/atomate2/forcefields/flows/phonons.py

[tpurcell90]

@tpurcell90 I was too curious to not look at it. I left several comments. One major comment that I have: The phonon forcefield workflow should probably also inherit from this new base class if possible: https://github.com/materialsproject/atomate2/blob/main/src/atomate2/forcefields/flows/phonons.py

The forcefields now have the same format as VASP/aims

[JaGeo]

The forcefields now have the same format as VASP/aims

@tpurcell90 Awesome! Thank you.

[JaGeo]

@utf prev_dir is now breaking the VASP workflow. Do you maybe have a good suggestion to solve this? Should we define prev_dir in addition to prev_vasp_dir or rename completely?

[tpurcell90]

Another option would be to add both now and give a deprecation warning for prev_vasp_dir to be removed at a later date.

[utf]

For now, can we follow the approach taken in the Magnetic ordering maker (see here) from this PR: #432

I will make a separate issue about renaming prev_vasp_dir which we can then implement after some consultation.

[tpurcell90]

I used that approach for the phonon workflows as well. I set the forcefields' argname to None since it does not look like that used a prev_dir keyword at all

[tpurcell90]

AimsInputSet is now in pymatgen master. So I need to wait for the next version to be released

[JaGeo]

@tpurcell90 You might likely run into this error. I added some hints here: https://github.com/materialsproject/atomate2/pull/693/files

[tpurcell90]

Thanks I'll take a look at those once I am back from class!

[tpurcell90]

So looking at this the pymatgen change is also in #693 so I think it makes more sense to wait for that to go in then deal with the merge conflicts of the two fixes.

[utf]

Hi @tpurcell90, are you able to resolve the remaining conflicts? Then this should be be mergeable.

[tpurcell90]

@utf This is now ready, let me know if there are any more final changes you want me to make

[utf]

Fantastic, thank you @ab5424 and @tpurcell90!