Fix: camelcase duplicate warning bug (fixes #10801)

[helixbass]

What is the purpose of this pull request? (put an "X" next to item)

[ ] Documentation update
[X] Bug fix (template)
[ ] New rule (template)
[ ] Changes an existing rule (template)
[ ] Add autofixing to a rule
[ ] Add a CLI option
[ ] Add something to the core
[ ] Other, please explain:

See #10801 for description of behavior

What changes did you make? (Give an overview)
Fixed the logic in lib/rules/camelcase.js, and added a test that would've been broken before the fix

Basically, there appears to have been faulty logic introduced by #10373:

This guard (pre-#10373) was preventing reporting the key of a Property:

node.parent.key === node && node.parent.value !== node

Then #10373 replaced that guard with this:

!assignmentKeyEqualsValue && node.parent.key === node

The hole introduced in the logic is if assignmentKeyEqualsValue but node.parent.key !== node.parent.value, the value gets reported twice (once while visiting the key and once while visiting the value)

I believe that the only reason existing invalid tests like var { category_id } = query; weren't failing on duplicate warnings is that the default parser (espree?) must reuse the same Identifier node for the key and value of a shorthand Property (as opposed to the Coffeescript custom parser I'm working on, which doesn't reuse the same node, thus how I found this bug)

But then by adding an invalid test for var { category_id: category_id } = query;, you get a duplicate warning (it exposes the hole in the logic, since assignmentKeyEqualsValue but the Property's key and value are two different Identifier objects)

So the fix is basically to restore the pre-#10373 guard logic so that you effectively never visit the Property key

Is there anything you'd like reviewers to focus on?

[platinumazure]

LGTM, thanks!

[platinumazure]

LGTM, thanks!

[kaicataldo]

Thanks for contributing to ESLint!