Amino Acid Geometry + Protein Builder

[ayushkarnawat]

For each of the amino acids, we currently define the geometries within each class. This leads to a lot of repetitive parameters being defined in each of the __init__ definitions.

Rather, we should have a geometry.json file that stores all the default geometry information for each amino acid. This way, we can have a single class that handles parsing the data so that the geometry information (i.e. bond lengths, angles, and dihedral angles) can be properly instantiated for each class.

[ayushkarnawat]

The implementation for building up the protein (from individual geometries) seems quite hacked together.

1. Inadequate documentation
2. No mathematical explanation for how new XYZ coordinates are computed

The following files/functions need to be refactored for robust performance:

1. Refactor how the Residue + Atom objects are created from the default geometries
2. Ability to both add and remove residues when creating PDB structure (currently we are only able to modify residues)

[ayushkarnawat]

Will not be updated/fixed in the near future, as we are now using PyMol's mutagenesis wizard (though the python wrapper) to mutate the sequences. See #9 for more details.