previously, `#eval` would happily evaluate expressions that contain
`sorry`, either explicitly or because of failing tactics. In conjunction
with operations like array access this can lead to the lean process
crashing, which isn't particularly great.
So how `#eval` will refuse to run code that (transitively) depends on
the `sorry` axiom (using the same code as `#print axioms`).
If the user really wants to run it, they can use `#eval!`.
Closes#1697
Many of our tests in `tests/lean/run/` produce output from `#eval` (or
`#check`) statements, that is then ignored.
This PR tries to capture all the useful output using `#guard_msgs`. I've
only done a cursory check that the output is still sane --- there is a
chance that some "unchecked" tests have already accumulated regressions
and this just cements them!
In the other direction, I did identify two rotten tests:
* a minor one in `setStructInstNotation.lean`, where a comment says `Set
Nat`, but `#check` actually prints `?_`. Weird?
* `CompilerProbe.lean` is generating empty output, apparently indicating
that something is broken, but I don't know the signficance of this file.
In any case, I'll ask about these elsewhere.
(This started by noticing that a recent `grind` test file had an
untested `trace_state`, and then got carried away.)
