Linear analysis added

[chrisiacovella]

This PR allows the director to be calculated not from the tails, but from a new region defined in molecules labeled "la_regions" (linear analysis regions). If la_regions isn't defined in the file, it will copy the indices from the tails section in the molecules file.

[codecov[bot]]

Codecov Report

Merging #52 into master will increase coverage by 1.17%. The diff coverage is 48.21%.

@@            Coverage Diff             @@
##           master      #52      +/-   ##
==========================================
+ Coverage   37.93%   39.10%   +1.17%     
==========================================
  Files          11       11              
  Lines         464      514      +50     
==========================================
+ Hits          176      201      +25     
- Misses        288      313      +25     

Impacted Files	Coverage Δ
analysis/load.py	15.10% <0.00%> (-0.69%)	:arrow_down:
analysis/utils.py	31.03% <0.00%> (ø)
analysis/residue.py	48.00% <44.44%> (-4.95%)	:arrow_down:
analysis/molecules/molecules.py	82.75% <62.16%> (-7.15%)	:arrow_down:

[chrisiacovella]

Looks like all checks are passing. Can you give this a once over to see if there are any issues I haven't accounted for?

[uppittu11]

I think this is okay for now. In the future it might be better to take a cleaner/extensible approach by making molecule.tails a dictionary so each tail has a name (i.e. {"sph": [list_of_indices], "fa": [list_of_indices], "fa_equal_tails": [list_of_indices]}). And then we can specify which tail(s) to use in the S2 calculation by name.

Also, currently the tilt angles in the results is an (N_frames*N_tails) array of the tilt angle. Labeling each column with the tail name would make it easy to filter them out by type.