contification paper

[Matthew Fluet]

> Matthew, based on my reading of the patent, I think we should go ahead with the
> paper.

Sounds good.  I've been trying to think of a good motivating example; the
nested loops from Reppy's paper is a simple one, but it would be nice to
have something that which isn't covered by the other analyses.  Maybe
something as simple as the even/odd pair of mutually recursive functions
would be good.  Something like:

fun fm () = ... even n ...
fun even n = ... odd n' ...
fun odd n = ... even n' ...

Only odd is contified by the call based analysis (although even would be
contified on the next contify simplification).  Nothing is contified by
the continuation based analysis.  But both even and odd are contified by
the dominator based analysis.