Add a ring opening/closing test case

[jchodera]

The null transformation of naphthalene to naphthalene (where a ring is destroyed at one end and created at the other end) would be a great test case since we know the free energy difference should be zero. (Jed Pitera had a paper on this test case with Wilfred van Gunsteren.). We can do this in vacuum, implicit solvent, and explicit solvent.

This would be a very powerful test because of the novelty of our ability to handle ring opening and closing.

[jchodera]

Jed's naphthalene paper is here: http://dx.doi.org/10.1080/08927020211973

[jchodera]

I've already implemented this test case.

[jchodera]

Whoops, my mistake: I only implemented benzene <-> napthalene, not the naphthalene null transformation (which would require specifying the atom mappings to use).

[pgrinaway]

Yeah, I almost made the same mistake and closed the issue, but you're right.

[pgrinaway]

These are in the repo.