iSpecialize and iDestruct.

These tactics now all take an iTrm, which is a tuple consisting of
a.) a lemma or name of a hypotheses b.) arguments to instantiate
c.) a specialization pattern.

These tactics are superfluous:
- iPure H as pat => iDestruct H as pat
- iPersistent H => iSpecialize H "!"

- It can now also frame under later.
- Better treatment of evars, it now won't end up in loops whenever the goal
  involves sub-formulas ?P and it trying to apply all framing rules eagerly.
- It no longer delta expands while framing.
- Better clean up of True sub-formulas after a successful frame. For example,
  framing "P" in "▷ ▷ P ★ Q" yields just "Q" instead of "▷ True ★ Q" or so.

Par is now defined as an expression of type [∀ X, expr X] (instead of a
value) and we prove that it is stable under weakening and substitution.

It is just a test case and not really part of the barrier library.

This gets rid of the (ambiguous) notation %l, because we can declare
LitLoc as a coercion. It also shortens the code.

That way, we do not have useless type annotations of the form
"v : language.val heap_lang" cluttering about any goal.

Note, that we could decide to eta expand everywhere (as we do for ∀
and ∃), and use the notation "WP e {{ Q }}" for "wp e ⊤ (λ _, Q)".

It is not a library; it does not contain code, but instead is a core
part of heap_lang.

This reverts commit 3cc38ff6.

The reverted pure hypotheses and variables appear in the wrong order.

Since strip_later is doing a good job stripping laters in the conclusion,
these tactics are thus no longer needed. Also, wp_finish now properly
converts the result in a primitive viewshift in case it is not a weakestpre.

and show that this is an instance of evaluation contexts