gh-94445: add compiler test for another case of excessive stack use

[carljm]

trafficstars

In GH-94329 the MAX_ALLOWED_STACK_SIZE check in the compiler (added in 3.10.0) was reverted, because it was discovered that a large sequence unpacking could use more than the allowed stack size. A todo comment was left in the code to re-add this check in future, and GH-94445 was created to track the todo item.

I've found another case in which unbounded stack can be consumed: a function with a very large number of type-annotated arguments. This PR simply adds a test for that case, to ensure that if in future the stack-size check is again enforced, the case won't be missed and cause another breakage for valid (if pathological) code.

[cpython-cla-bot[bot]]

All commit authors signed the Contributor License Agreement.

[miss-islington]

Thanks @carljm for the PR, and @iritkatriel for merging it 🌮🎉.. I'm working now to backport this PR to: 3.11. 🐍🍒⛏🤖

[bedevere-bot]

GH-99262 is a backport of this pull request to the 3.11 branch.