Generalize dilute elyte model for arbitrary binary electrolyte

[Raymond Smith]

The code currently implicitly assumes z_+ = -z_- = nu_+ = nu_- = 1, which is not good. One option is to simply throw errors when the above are not satisfied and the dilute model is requested. However, the equations aren't that different, and this could be done with issue #8.

[Raymond Smith]

In fact, while we're at it, why not just forgo all the silly rearrangement into Damb and transference numbers and just simulate the more basic:

1) conservation of anion (to avoid a reaction term in this equation)

2) conservation of charge

[Edwin Khoo]

I agree with that. It also allows the model to be more easily extended to multiple cations and anions (so there isn't a binary electrolyte to even define Damb).

[Raymond Smith]

Oh, good point about non-binary. Yeah, totally agree.

[Raymond Smith]

Merged in elyte-refactor-and-generalize (pull request #36)

Elyte refactor and generalize, Fix issue #33, see issue #8

→ <<cset b216172bf214>>