Refactor double well bond potential

[mphoward]

@clpetix Can you polish this one off with documentation when you get a chance?

[clpetix]

On it!

[astatt]

@clpetix I apologize for creating more work for you! After talking to my student (the user of this potential), I think it makes the most sense to define:

• location of the first minimum (a/2-b)

• a/2 (location of the transition state)

  instead of a and b, i.e redefining the variables a bit.

Then I suggest the following naming/documentation:

• V_barrier Potential barrier height between the two minima as measured between the first minimum and maximum
• r_tr location of the transition state where the potential barrier is maximum
• r_0 location of the first minimum
• V_offset Potential height in the activated state with respect to the first minimum

Then, in the documentation we can mention the equation for the location of the second minimum etc.

Does this sound good and makes sense? Feel free to ping me if you need input/help!

[clpetix]

We updated the unit tests to make sense. The new implementation means that the differences of r, r0, and r1 could all be shifted arbitrarily without messing with the energies or forces.

This should be ready for y'all to review @astatt @mphoward !