Introduce Basic Orientational Entropy Calculations

[skfegan]

Summary

This pull request adds the basic orientational entropy calculation.

Changes

Adding Neighbours to Levels:

• Functions for counting the number of neighbours (two methods available - RAD and grid)
• Function for guessing symmetry number using RDKit structure matching functions
• Function for identifying linear molecules using RDKit labels of hybridization types for molecules with more than two united atom beads

Orientational Entropy Calculation:

• Adds the orientational term to the entropy calculation

Conformational Entropy:

• Fixing a small bug in conformational entropy by removing one equals sign in the identifying of peaks.

Documentation

• Update to science.rst to describe the orientational entropy calculation
• Update to conf.rst to include the neighbours search_type option

Impact

• The orientational entropy term is now included to give a more complete entropy value.
• The conformational entropy is improved by not identifying extraneous peaks in the histogram.
• Documentation expanded to cover the orientational entropy theory.

[harryswift01]

This addition to CodeEntropy is a major milestone in bringing new theory into the package, and this PR communicates that work very well. I’ve requested a few changes, but they are mainly focused on tidying up comments and clarifying certain areas. Functionally, this is looking very strong.

Since this introduces new entropy calculations, we should expect the regression test results to change. For that reason, I think it would be appropriate to re-run and update the baseline results to reflect the new calculation method. These can be updated locally with: pytest tests/regression --update-baselines. It also looks like coverage has dropped slightly. It would be great if we could bring that back up before merging.

We’re very close to getting this merged, thank you. If there’s anything I can do to help move this across the line, just let me know.