Attribute panics to the mdtests that cause them

[dcreager]

This updates the mdtest harness to catch any panics that occur during type checking, and to display the panic message as an mdtest failure. (We don't know which specific line causes the failure, so we attribute panics to the first line of the test case.)

Sample output (for an artificial panic):

---- mdtest__unary_integers stdout ----

integers.md - Unary Operations - Unary Addition

  crates/red_knot_python_semantic/resources/mdtest/unary/integers.md:6 panicked at crates/red_knot_python_semantic/src/types/infer.rs:3264:66:
  crates/red_knot_python_semantic/resources/mdtest/unary/integers.md:6 bonk

To rerun this specific test, set the environment variable: MDTEST_TEST_FILTER="integers.md - Unary Operations - Unary Addition"
MDTEST_TEST_FILTER="integers.md - Unary Operations - Unary Addition" cargo test -p red_knot_python_semantic --test mdtest -- mdtest__unary_integers

integers.md - Unary Operations - Unary Subtraction

  crates/red_knot_python_semantic/resources/mdtest/unary/integers.md:14 panicked at crates/red_knot_python_semantic/src/types/infer.rs:3265:66:
  crates/red_knot_python_semantic/resources/mdtest/unary/integers.md:14 foobar

To rerun this specific test, set the environment variable: MDTEST_TEST_FILTER="integers.md - Unary Operations - Unary Subtraction"
MDTEST_TEST_FILTER="integers.md - Unary Operations - Unary Subtraction" cargo test -p red_knot_python_semantic --test mdtest -- mdtest__unary_integers

Closes #13899

[dcreager]

All of the chatter in this crate is just to publicize this existing helper function.

[dcreager]

This does not include the full backtrace of the panic, since that seemed to be more verbose than we need at this point. If we want it, we just change info.info to info here.

[github-actions]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

✅ ecosystem check detected no linter changes.

[AlexWaygood]

Nice, this looks great! :D

[MichaReiser]

Neat

[AlexWaygood]

Hmm, this PR unfortunately now seems to be suppressing some of the internal error messages from red_knot_test, for example this one here:

ruff/crates/red_knot_test/src/lib.rs

Lines 32 to 37 in b264489

	let suite = match test_parser::parse(short_title, &source) {
	Ok(suite) => suite,
	Err(err) => {
	panic!("Error parsing `{path}`: {err:?}")
	}
	};

On a PR branch I'm working on, I got this (not very informative!) test failure:

If I revert this commit on that PR branch, I get this instead, which is much more helpful:

[AlexWaygood]

I opened #15317 so we don't forget about this 👍

[dcreager]

Thanks Alex! I'll take a look