Generalize workflow to be better able to work with coarse graining

[jennyfothergill]

right now planckton can initialize atomistic opv sims, but how can we make it easier to run sims of CG systems?

Big picture ideal workflow:

• make a polymer system using smiles for the beads (using planckton?)
• make a CG mapping and model (using grits or chris' polymer utils)
• calculate a ff for the CG model and run it (using MSIBI)
• backmap to atomistic with the CG mapping (using grits)
• calculate charge mobility (using morphct)

For planckton to fit in this workflow it needs to...

• separate initialization from typing
• make it easier to provide a custom ff