Support CIF files from ICSD with atom type symbols like Au0+

[uellue]

A recent CIF file pulled from https://icsd.fiz-karlsruhe.de/ for collection code 163723 contained the atom type symbol "Au0+" instead of just Au. This lead to empty diffraction patterns.

With this change only the first alphabetical
part of the atom type symbol is used as an element, and a warning is emitted if unknown elements are encountered.

See https://www.iucr.org/__data/iucr/cifdic_html/1/cif_core.dic/Iatom_type_symbol.html for the definition of the atom type symbol.

It can be debated if unrecognized atom type symbols should actually be a hard error since the current
code will quietly zero the contribution of sites with unknown atom types, which may lead to plausible but wrong diffraction patterns if recognized and unrecognized atom type symbols are mixed in a CIF file.

Description of the change

Progress of the PR

• Docstrings for all functions
• Unit tests with pytest for all lines
• Clean code style by running black

Minimal example of the bug fix or new feature

EntryWithCollCode163723.cif.txt

from diffpy.structure import loadStructure
from diffsims.generators.simulation_generator import SimulationGenerator
from orix.quaternion import Rotation
from orix.crystal_map import Phase
gen = SimulationGenerator(
    accelerating_voltage=300,
)
rot = Rotation.from_axes_angles(
    [1, 0, 0], 0, degrees=True
)
structure = loadStructure('EntryWithCollCode163723.cif')
p = Phase(structure=structure, space_group=225)
sim = gen.calculate_diffraction2d(phase=p, rotation=rot)
# Previously gave you an empty list
print(sim.coordinates)

For reviewers

• The PR title is short, concise, and will make sense 1 year later.
• New functions are imported in corresponding __init__.py.
• New features, API changes, and deprecations are mentioned in the
  unreleased section in CHANGELOG.rst.
• Contributor(s) are listed correctly in credits in diffsims/release_info.py and
  in .zenodo.json.

[uellue]

pre-commit.ci autofix

[uellue]

The docs failure seems to be an issue with Sphinx or an extension?

[hakonanes]

Good fix, @uellue!

The doc fail is related to pyxem/orix#570, I think.

@CSSFrancis, how about a patch release v0.7.1 with this fix? Nothing on main since v0.7.0.

[CSSFrancis]

@hakonanes Sounds good to me :)

[uellue]

About __init__.py I am unsure: Should get_element() be included there?

[uellue]

Is there anything that I can still do here? @hakonanes @CSSFrancis

[CSSFrancis]

@uellue let me try to find some time to look at this today!

[CSSFrancis]

@uellue looks good to me! I'll let the CI run and then merge it.

By the way i did get somewhat far on adding ionic scattering cross-sections #222