Consider extending `suppressible-exception` (`SIM105`) to multi-line `try` blocks

[DetachHead]

try: # error
    ...
except Exception:
    pass

try: # no error
    ...
    ...
except Exception:
    pass

playground

[KotlinIsland]

I think this is intentional, although I don't really see why.

[konstin]

For context, this is due to #1947/#1948 and flake8-simplify doing the same (https://github.com/MartinThoma/flake8-simplify/blob/60e0ddbc82973af673259850cc00c6e8221f8643/flake8_simplify/rules/ast_try.py#L46). I'm not sure if that's the best way to handle this though

[charliermarsh]

I think the current restriction is not unreasonable, but I'd be open to reviewing a PR that modifies the rule to use the more limited exemption from #1947:

If a try: block has an unreachable end because all its non-exceptional control flow paths end with a return, raise, break, or continue, should we exempt it from SIM105?

Though it'd be contingent on reviewing the changes in the ecosystem checks.

[DetachHead]

workaround: use S110 instead. though it has a different error message, it seems to be pretty much the same rule except it also covers this case as well.

perhaps SIM105 should be deprecated in favor of S110 and S110's error message can be updated to also mention contextlib.suppress?