Improve invalid-slice-index and add invalid-slice-step

[cdce8p]

Description

Followup to pylint-dev/astroid#1844 (comment) by @nelfin.

This PR

• Improve the invalid-slice-index check to detect invalid indices on string and byte sequences.

"Hallo"["a":"z"]  # invalid-slice-index (2x)

• Adds a new warning for invalid-slice-step when 0 is passed as step value for common builtin sequences.

["a", "b", "c"][::0]  # invalid-slice-step

[coveralls]

Pull Request Test Coverage Report for Build 3457177074

• 9 of 10 (90.0%) changed or added relevant lines in 1 file are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage increased (+0.0008%) to 95.405%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
pylint/checkers/typecheck.py	9	10	90.0%

Files with Coverage Reduction	New Missed Lines	%
pylint/checkers/typecheck.py	1	96.05%

Totals
Change from base Build 3455918345:	0.0008%
Covered Lines:	17336
Relevant Lines:	18171

---

💛 - Coveralls

[Pierre-Sassoulas]

LGTM, the error level check are always a pleasure to review :) (Line not covered is due to inference it's hard to test in the current state of the testutil)

[cdce8p]

(Line not covered is due to inference it's hard to test in the current state of the testutil)

It's actually a bit more complicated. _check_invalid_slice_index is only called from _check_invalid_sequence_index which includes a pretty strong filter of things too check. So all cases (with custom __getitem__ methods) don't even get to the slice_index check.

I wouldn't want to remove the uncovered line in _check_invalid_slice_index though, as at some point we might want to loosen the strict guard clause a bit.

https://github.com/PyCQA/pylint/blob/435a5a1d8e542d1240b8a66985feb8950fb6d739/pylint/checkers/typecheck.py#L1695-L1701

[github-actions]

🤖 According to the primer, this change has no effect on the checked open source code. 🤖🎉

This comment was generated for commit f0638f8