Improved creation and handling of sets of reciprocal lattice vectors

[hakonanes]

Description of the change

This PR introduces a new class ReciprocalLatticeVector, replacing ReciprocalLatticePoint for creation and handling of sets of reciprocal lattice vectors hk(i)l. The latter class only supported cubic phases, while the new class supports all crystallographic point groups. It does so by relying on orix for all (symmetrically reduced) vector handling via Miller, while not directly extending that class since the new class does not support direct lattice vectors. A new class is necessary, instead of just using Miller, for easy calculation of structure factors and Bragg angles. This class will be used in kikuchipy for (existing) geometrical and (to come) kinematical EBSD simulations.

This is the last thing I'd like to get released in v0.5.

Closes #144.

Progress of the PR

• [x] Docstrings for all functions
• [x] Unit tests with pytest for all lines
• [x] Clean code style by running black
• [ ] ~User guide notebook detailing use of the new class~ Another time, perhaps
• [x] Complete API reference
• [x] Changelog entry
• [x] Add a deprecation warning to ReciprocalLatticePoint

Minimal example of the bug fix or new feature

>>> from orix.crystal_map import Phase
>>> from diffsims.crystallography import ReciprocalLatticeVector
>>> phase = Phase.from_cif("al.cif")
>>> phase
<name: al. space group: Fm-3m. point group: m-3m. proper point group: 432. color: tab:blue>
>>> phase.structure
[Al   0.000000 0.000000 0.000000 1.0000,
 Al   0.000000 0.500000 0.500000 1.0000,
 Al   0.500000 0.000000 0.500000 1.0000,
 Al   0.500000 0.500000 0.000000 1.0000]
>>> phase.structure.lattice
Lattice(a=4.04958, b=4.04958, c=4.04958, alpha=90, beta=90, gamma=90)
>>> ref = ReciprocalLatticeVector.from_min_dspacing(phase, 0.7)
>>> ref[:10]
ReciprocalLatticeVector (10,), al (m-3m)
[[ 5.  2.  2.]
 [ 5.  2.  1.]
 [ 5.  2.  0.]
 [ 5.  2. -1.]
 [ 5.  2. -2.]
 [ 5.  1.  2.]
 [ 5.  1.  1.]
 [ 5.  1.  0.]
 [ 5.  1. -1.]
 [ 5.  1. -2.]]
>>> ref.calculate_structure_factor("xtables")
>>> ref.calculate_theta(20e3)
>>> ref = ref[ref.structure_factor.real > 0]
>>> ref.print_table()
 h k l      d       I     |F|_hkl  I_Rel.   Mult 
 1 1 1    2.338    72.1     8.5    100.0     8   
 2 0 0    2.025    50.0     7.1     69.3     6   
 2 2 0    1.432    19.8     4.4     27.4     12  
 3 1 1    1.221    13.1     3.6     18.2     24  
 2 2 2    1.169    11.8     3.4     16.3     8   
 4 0 0    1.012    8.1      2.9     11.3     6   
 3 3 1    0.929    6.5      2.5     9.0      24  
 4 2 0    0.906    6.0      2.5     8.4      24  
 4 2 2    0.827    4.7      2.2     6.5      24  
 3 3 3    0.779    4.0      2.0     5.5      8   
 5 1 1    0.779    4.0      2.0     5.5      24  
 4 4 0    0.716    3.1      1.8     4.4      12  
>>> fig = ref.draw_circle(axes_labels=["a", "b"], opening_angle=np.pi / 2 - ref.theta, return_figure=True, linewidth=0.75)
>>> ref.draw_circle(opening_angle=np.pi / 2 + ref.theta, figure=fig, linewidth=0.75)

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.

[hakonanes]

There are many line additions in this PR, however:

• many of them are docstring examples
• much of the functionality is duplicated from the old ReciprocalLatticePoint, which is deprecated in this PR in favour of ReciprocalLatticeVector

I hope the docstring examples and inline comments are sufficient to understand the code.

EDIT: Generation of sets of vectors and calculation of structure factors have been compared for cubic, hexagonal and monoclinic phases to the Bruker DynamicS software.

EDIT 2: See the reference docs generated from this PR for an overview of the new class.

[hakonanes]

No pyxem tests in the current master branch fail when run on Ubuntu with this PR's version of diffsims.

[pc494]

I'll be able to review this tomorrow, so we can probably get 0.5.0 out next week?

[hakonanes]

That would be fantastic, thank you Phillip. If this is merged tomorrow I suggest we release 0.5.0 right away.

[pc494]

Tomorrow works for me, will try and get my review in early in that case

[hakonanes]

Fantastic, thank you a lot, Phillip. I'll update the release date in the changelog in #187, then we can merge that and publish 0.5.0.