[optional ext] Emit redefined-loop-name for redefinitions of loop variables in body

[jacobtylerwalls]

Type of Changes

	Type
✓	✨ New feature

Description

Closes #5072 -- Adds new message in optional extension
Closes #5608 -- Uncouples the nested loop redefinition case from redefined-outer-name so the check can be enabled/disabled appropriately

# 1
for y in iterable:
    for y in iterable:  # [already emitted redefined-outer-name, now emits redefined-loop-name]
        ...

# 2
for y in iterable:
    y = y.lower()  # [with this PR emits redefined-loop-name]

Ultimately I'm open to the idea of making this into a new message, but then we should get a decision on whether to bring example 1 into the new message also. done!

[coveralls]

Pull Request Test Coverage Report for Build 2255071021

• 47 of 47 (100.0%) changed or added relevant lines in 3 files are covered.
• 116 unchanged lines in 11 files lost coverage.
• Overall coverage increased (+0.04%) to 95.162%

---

Files with Coverage Reduction	New Missed Lines	%
pylint/testutils/checker_test_case.py	1	97.96%
pylint/testutils/pyreverse.py	1	97.92%
pylint/checkers/classes/special_methods_checker.py	2	94.83%
pylint/config/arguments_manager.py	2	98.13%
pylint/reporters/base_reporter.py	2	92.11%
pylint/checkers/utils.py	8	95.78%
pylint/reporters/text.py	9	85.61%
pylint/testutils/lint_module_test.py	10	86.67%
pylint/checkers/variables.py	23	96.61%
pylint/lint/pylinter.py	26	94.92%

Totals
Change from base Build 2235025626:	0.04%
Covered Lines:	15814
Relevant Lines:	16618

---

💛 - Coveralls

[Pierre-Sassoulas]

I'm not very clear on the best way between extending redefined outer name and creating a new checker. We could create a new message without creating a new checker too. I hope for a consensus to be reached in the discussion.

[Pierre-Sassoulas]

Look pretty good already !

[DanielNoord]

@jacobtylerwalls We might need to take a look at the failing pylint run here.

[Pierre-Sassoulas]

Outch, if we have that much new warning ourselves I guess it's going to surprise users too. I'm more inclined to create a new optional checker now.

[jacobtylerwalls]

Yes, I'm inclined toward that also, but I really do think we need to ask whether #1 from the top of the thread should become part of the new checker then.

[jacobtylerwalls]

@superbobry do you have thoughts on this approach? This would allow you to disable redefined-loop-name and also be disabling the nested loop case, uncoupling it from redefined-outer-name.

Also, do you have thoughts on making this optional or default? I will probably disable this in my own projects.

[jacobtylerwalls]

Updated to make this an optional extension.

[jacobtylerwalls]

Failing docs check is not relevant, see #6470

[DanielNoord]

As I said, I really like the test. Sorry for letting this sit so long, that causes a number of change-requests due to my own aggressive deprecation PRs related to argparse and mypy 😅 Sorry about that!

[Pierre-Sassoulas]

I like the creation of a new checker from the VariableChecker. Some checks definitely need to be disentangled to make maintenance easier.