translate ripken-wolski-forest coupling to sagan

[mattsignorelli]

We should have a feature which translates the B,E,K,H matrices of RWF to the Sagan formalism in current Bmad

[mattsignorelli]

The Sagan formalism should be the default because it is simpler and works for most lattices, but users should be able to use the more general RWF formalism when needed

[mattsignorelli]

In Ripken-Wolski-Forest:

And in Sagan;

So we can trivially get V from H, and then compute the Sagan lattice functions from

We should do it this way so that we can properly be on the parameter-dependent fixed point