[Php83] Allow indirect override abstract method on Grand Child on AddOverrideAttributeToOverriddenMethodsRector

[samsonasik]

continue of patch:

• [Php83] Skip add #[\Override] on non-empty method from trait on AddOverrideAttributeToOverriddenMethodsRector #6471

abstract method override can be layered children override, so indirect override should be be make sense to override.

see https://getrector.com/demo/45456834-d91c-4036-aa63-3d920562498d

the GrandChild class should be allowed have make sense to have #[\Override]

[samsonasik]

All checks have passed 🎉 @TomasVotruba I am merging it ;)

/cc @PhilETaylor please test this in case of regression for this change. This basically allow add #[\Override] on indirect override abstract method, eg:

abstract class SomeAbstractClass
{
    public function run();
}

class SomeChild extends SomeAbstractClass
{
    public function run()
    {
    	echo 'default';
    }
}


class GrandChild extends SomeChild
{
+   #[\Override]
    public function run()
    {
    	echo 'override default';
    }
}

The GrandChild should have added #[\Override] as override non-empty method from its parent class. The first override from abstract method doesn't have #[\Override], but the child need have as parent have content, and it override it.

[PhilETaylor]

  - Downloading rector/rector (dev-main 9e165cc)
  - Upgrading rector/rector (dev-main 87ea771 => dev-main 9e165cc): Extracting archive

Ran my cleanup and nothing changed, so no regression on my specific code like there was last week. However, my issue last week was with a class using a trait, and some methods in that class were incorrectly getting the Override attribute for some methods. You fixed that last week.

On my main project I run bleeding edge dev-main rector, update composer deps daily, and run cleanup (ecs, rector etc) daily, but when things go wrong I dont ask for support but if I can replicate I like to report bugs.

[samsonasik]

Thank you for verify 👍