Fix regression in the creation of the __init__ of dataclasses

[DanielNoord]

Steps

• For new features or bug fixes, add a ChangeLog entry describing what your PR does.
• Write a good description on what the PR does.

Description

Closes pylint-dev/pylint#7434

Type of Changes

	Type
✓	🐛 Bug fix

[coveralls]

Pull Request Test Coverage Report for Build 3176405357

• 57 of 74 (77.03%) changed or added relevant lines in 2 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.1%) to 92.295%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
astroid/brain/brain_dataclasses.py	27	29	93.1%
astroid/nodes/node_classes.py	30	45	66.67%

Files with Coverage Reduction	New Missed Lines	%
astroid/nodes/node_classes.py	1	94.78%

Totals
Change from base Build 3151597495:	-0.1%
Covered Lines:	9859
Relevant Lines:	10682

---

💛 - Coveralls

[DanielNoord]

I expected the coverage to be sup-optimal. I think it's okay to merge this right now. If we start using this method more then the coverage (and potential bugs) will come naturally. You just don't have *args and **kwargs in dataclasses.

[Pierre-Sassoulas]

I'm way over my head with this MR, I'm going to pass unless there's really no other reviewer.

[jacobtylerwalls]

Just an initial skim.

[jacobtylerwalls]

Can you explain this change? If it's harmless, then consider not making the change to reduce the diff. The elif/else branches below also have the same trailing separator.

[DanielNoord]

prev_kw_only now takes care of this itself in the new helper function so there is not need to add it here anymore.

Edit: Let me know if you want me to change this. Otherwise I think this can be merged.

[jacobtylerwalls]

Thanks, that explains it. Merge ho!