Lattice Usage

[PaulWAyers]

@wilhadams can you help explain how the Lattices work? I think we need some more top-level documentation here.
@ZhaoYilin can you look at the lattices and see if it is sensible?

I am not sure what we support right now..... If we support triclinic then we need the ability to define several different types of parameters due to inequivalent bonds (even if all sites are equivalent). If we have red/black (different site-types) or defect support it gets even more complicated.

[PaulWAyers]

@RichRick1 it may be worth making this an issue. It would be good to have specifying a lattice connectivity/geometry as an alternative to an adjacency matrix. (Essentially a front-end to the adjacency matrix.) I can imagine how to do it with red-black or red-black-blue lattices, where one puts in "dummy atom types" in a connectivity pattern (using the atom-type adjacency specification).