Get default feature ranges based on conventional cells

[sgbaird]

Follow-up to #36. The idea is to add to https://github.com/sparks-baird/xtal2png/blob/main/notebooks/2.0-materials-project-feature-ranges.ipynb or make a copy and modify it so the structures are reduced to their primitive conventional representations.

 spa = SpacegroupAnalyzer(structure, symprec=0.1, angle_tolerance=5.0) 
# structure = spa.get_primitive_standard_structure()
structure = spa.get_conventional_standard_structure()

@hasan-sayeed you OK with tackling this? Should be pretty straightforward, just splicing the above code into a copy of the linked notebook.

[sgbaird]

For visibility, here's some of the summary data from the nb

Ranges (Min/Max)

Conventional

range of a is:  1.648906 - 72.590284
range of b is:  2.263836 - 83.00690244
range of c is:  2.131537 - 194.82034183895985
range of volume is:  15.216262193734085 - 56652.78402596729
range of num_sites is:  2 - 208
range of pair-wise distance is:  0.7249349602879995 - 97.40907797736222

Primitive

range of a is:  2.296021 - 66.29136774227022
range of b is:  2.258778 - 61.125585795588215
range of c is:  2.131537 - 130.453537
range of volume is:  11.91856931582488 - 20090.90640762975
range of num_sites is:  2 - 52
range of pair-wise distance is:  0.7249349602879995 - 64.8913973530744

Upper 0.99 Quantile

Conventional

a                 18.875420
b                 18.594637
c                 39.051768
volume          2800.497775
num_sites        114.000000
max_distance      19.332560

percent retained TBD

Primitive

a                 15.292415
b                 14.953414
c                 35.792380
volume          1467.529411
num_sites         52.000000
max_distance      17.550941

'a < @upp_df.a and b < @upp_df.b and c < @upp_df.c and volume < @upp_df.volume and num_sites < @upp_df.num_sites and max_distance < @upp_df.max_distance'

95.7% retained

Using conventional cell (e.g. via encode_as_conventional, decode_as_conventional flags, or more simply via encode_structure_type or similar) as a hyperparameter might make sense in the context of #79.