This is very experimental. It should now deal better with stuff like:

  match x with .. end = match y with .. end

In case there is a hypothesis H : R x y, it will try to destruct it.

This is more consistent with the proofmode, where we also call it pure.

This is more natural, match should be used in user code, not case.

Fun test case to play arround with abstract predicates.

Fixes issue #20.

This is to avoid confusion with ghost_ownership.own.

This introduces n hypotheses and destructs the nth one.

This is inspired by ssr, and makes unification faster if it goes
right-to-left.

See https://sympa.inria.fr/sympa/arc/ssreflect/2013-11/msg00010.html