As reported by @jjourdan: framing now no longer back tracks on whether
to strip laters or not. When framing below a later, we now only make it
strip laters of the head of the frame.

Remove the domain finiteness hypothesis for the function CMRA, and put cmra_extend in Type.

See merge request FP/iris-coq!118

The finiteness was needed to have the axiom of choice over the domain. This axiom is not needed if cmra_extend is in Type.

Revert "Remove the domain finiteness hypothesis for the function CMRA, and put cmra_extend in Type."

This reverts commit fa897ff5.

The finiteness was needed to have the axiom of choice over the domain. This axiom is not needed if cmra_extend is in Type.

Stronger `iNext` that performs arithmetic cancelation

Closes #148

See merge request FP/iris-coq!109

Fixed by stdpp 93b4ec70e13a573a9055a5bf1269f5885e18e843.

We now use the `Maybe` prefix as also used for `Frame`: it indicates
whether progress has been made by stripping of a later or not.

Use `NoBackTrack` type class for framing with ▷

Closes #153

See merge request FP/iris-coq!112

Various improvements to iFrame

Closes #145

See merge request FP/iris-coq!110

We already supported framing under wands.

For example, framing `P` in `(P ∨ Q) ∗ R` now succeeds and turns the goal into `R`.