Stabilize `PLR0917`

[flying-sheep]

I think it looks pretty stable, no?

[ntBre]

I think it does look pretty stable. Searching for related issues and PRs turns up a lot of results, but the ones I checked only referred to PLR0917 and didn't actually modify it. We should consider stabilizing it in the next minor release.

[MichaReiser]

Trying to remember what the reason was that we didn't stabilize the rules...

The only thing that comes to my mind is that we had concerns around various complexity rules (this is a one-off because it tries to approximate complexity by the number of positional arguments). That would explain why many of the other too-many rules haven't been stabilized. But I wasn't able to find any mention in notion. @AlexWaygood do you remember why we haven't stabilized this rule (it was added in 2023)

[ntBre]

I'm looking at this again for 0.13, and I think there are a couple of potential blockers for stabilization:

• It's a pretty opinionated rule, so we may want to wait for Rule categorization #1774 (I commented this here too)
• We may want to update the rule to skip pytest hooks (FBT001 and PLR0917 - false positive on pytest functions #7286 (comment))

Related to the first point, clippy's analogous rule has a default severity of warn, so #1256 may be related too.

I think we should probably resolve the second issue before stabilization, but I don't think being opinionated is necessarily a hard blocker on its own.

[flying-sheep]

PLR0913 (too-many-arguments) used to work like this one and was changed to be less useful, this is why this rule exists. This rule has a strictly more narrow scope as too-many-arguments, so it’s also strictly less opinionated.

[flying-sheep]

Damn, I should have pushed this more, now 0.13 has been released without it