[flake8-simplify] Apply SIM113 when index variable is of type int

[njhearp]

Summary

Fixes #21393

Now the rule checks if the index variable is initialized as an int type rather than only flagging if the index variable is initialized to 0. I used ResolvedPythonType to check if the index variable is an int type.

Test Plan

Updated snapshot test for SIM113.

[astral-sh-bot]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

ℹ️ ecosystem check detected linter changes. (+3 -0 violations, +0 -0 fixes in 2 projects; 53 projects unchanged)

rotki/rotki (+1 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

+ rotkehlchen/tests/unit/decoders/test_dripsv1.py:79:9: SIM113 Use `enumerate()` for index variable `idx` in `for` loop

zulip/zulip (+2 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview --select ALL

+ zerver/lib/export.py:1962:9: SIM113 Use `enumerate()` for index variable `dump_file_id` in `for` loop
+ zerver/lib/export.py:2892:9: SIM113 Use `enumerate()` for index variable `dump_file_id` in `for` loop

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
SIM113	3	3	0	0	0

[ntBre]

Thank you!

Would you mind making this a preview change? It seems to be in the spirit of the rule and also matches the upstream linter for this test case, but I see there are some ecosystem hits.

Preview will also be nice in case we want to iterate further by inferring the type of the argument, for example with is_int. I think ResolvedPythonType only handles literals and simple operations on them like 1 + 2 rather than using type annotations. (I'm not saying we need to or should do that, just an idea)

[dscorbett]

SIM113’s recommendation is only safe when the index variable’s initial value is a strict instance of int, whereas something annotated int could be an instance of e.g. bool or a custom subclass, which would make the recommendation unsafe because the program’s behavior could change.

[ntBre]

Ah, thanks for the clarification! Disregard my last paragraph then, but I'd still lean toward making it a preview change.

[njhearp]

Thanks for the feedback! I updated the rule to be gated behind preview for strict instances of int. I also updated the documentation and comments. It looks like some actions failed due to timeouts, so I suspect some GitHub services are temporarily down and they will pass later.

[ntBre]

Thank you!