For consistency's sake.

This cleans up some ad-hoc stuff and prepares for a generalization
of saved propositions.

It now also contains a non-expansiveness proof.

I have simplified the following CMRA axioms:

  cmra_unit_preservingN n x y : x ≼{n} y → unit x ≼{n} unit y;
  cmra_op_minus n x y : x ≼{n} y → x ⋅ y ⩪ x ≡{n}≡ y;

By dropping off the step-index, so into:

  cmra_unit_preservingN x y : x ≼ y → unit x ≼ unit y;
  cmra_op_minus x y : x ≼ y → x ⋅ y ⩪ x ≡ y;

The old axioms can be derived.

This replaces f_equiv and solve_proper with our own, hopefully better, versions

The performance gain seems neglectable, unfortunatelly...

* Use sig instead of sigT: the proof is a Prop after all
* Tweak implicit arguments
* Shorten proof of sigma

It now traverses terms at most once, whereas the setoid_rewrite
approach was travering terms many times. Also, the tactic can now
be extended by defining type class instances.

This way it behaves better for discrete CMRAs.

This better seals off their definition. Although it did not give
much of a speedup, I think it is conceptually nicer.

I am now also using reification to obtain the indexes corresponding to
the stuff we want to cancel instead of relying on matching using Ltac.

With Set Printing All, these notations make me loose overview entirely.