Fix e_factor implementation

Right now, the e_factor variable isn't actually doing anything. Based on how its used in planckton, I think all that is happening is that it is scaling the non-bonded LJ params. If this is the case, then we can iterate over the parmed object and scale them there. Need to double check first.

Locate the e_factor implementation and compare its use with planckton, as described in the issue. Verify whether it should scale non-bonded Lennard-Jones parameters on the ParmEd object; done means the confirmed behavior is implemented and checked against the existing behavior.

Written by the indexing model from the issue text.