GH-15994: fix suggestion that anonymous classes be made abstract

[DanielEScherzer]

In the process, remove the (incorrect) assumption that any abstract method that
needs to be implemented by a class that cannot itself be made abstract must be
a private method - the existing test for an enum already showed that this was
not the case.

[DanielEScherzer]

The first commit is already sent at #15993, but to avoid merge conflicts I sent this as a follow-up to that patch - marking this as a draft until that merges

[DanielEScherzer]

@iluuu1994 would you mind taking a look? This is a follow-up to #15993

[iluuu1994]

Hmm, this message doesn't make too much sense to me, as we cannot implement a method that is also declared as abstract. Maybe we should disallow abstract functions in anonymous classes in the compiler. This message would still make sense when the method comes from a parent class, interface or trait.

Granted, that is not really related to this change, so I'm happy if you move the abstract method to a parent class.

In reality, we should probably also change the error message for this:

class Foo {
    public abstract function bar();
}

Fatal error: Class Foo contains 1 abstract method and must therefore be declared abstract or implement the remaining methods (Foo::bar)

Implementing Foo::bar() is a weird suggestion, given the abstract declaration comes from this class.

[DanielEScherzer]

Maybe we should disallow abstract functions in anonymous classes in the compiler.

I can work on a patch for that

I'm happy if you move the abstract method to a parent class.

done

[iluuu1994]

LGTM, I'll merge as soon as tests have passed 🙂

[iluuu1994]

Thank you!