clarify fractional bondorder method support

[mikemhenry]

See #41 (comment) for some details.

[mattwthompson]

I'm in favor of this change as it makes it clear what is and is not supported.

I don't think the scope is large enough to necessitate an EP review, but I do think the change is not strictly negligible and I'm not going to approve + merge it alone. I'd like input from @j-wags (or possibly @lilyminium) if implementing this would impact any fitting efforts. It's possible I'm unaware of experiments using other methods - since, well, I'm not one of the people doing them. The current implementation defines a default but does not seem to forbid other values from being passed through.