Change non-monotonic AM predictor coefficients

[wthrowe]

Either version works, but this version makes the analytic expressions nicer and is what will be in the paper.

Note that this does not affect the BBH paper, because that uses the monotonic method.

Proposed changes

Upgrade instructions

Code review checklist

• The code is documented and the documentation renders correctly. Run
  make doc to generate the documentation locally into BUILD_DIR/docs/html.
  Then open index.html.
• The code follows the stylistic and code quality guidelines listed in the
  code review guide.
• The PR lists upgrade instructions and is labeled bugfix or
  new feature if appropriate.

Further comments

[nilsdeppe]

@wthrowe I don't know if this relates at all to your group meeting update, but in any case, I'll let you merge it when you are ready :)

[wthrowe]

That's a good question, actually. Maybe I should hold off on this until I figure out exactly what we want.

[knelli2]

@nilsdeppe @wthrowe Maybe post a quick synopsis of what was brought up so we have a record of it?

[wthrowe]

Basically, what's on develop, what's in this PR, and what I wrote in my paper draft are all different. They all satisfy the requirements on the method, so it's not clear which one to choose. (And I'm now also considering a fourth version.)

[wthrowe]

Decided to go with a simpler version without the conditional on alignment. This should be ready now.