[KSP2] Resolver#overrides() gives incorrect result when method isn't accessible.

[bcorso]

Consider the following two classes:

package a;
public class A {
  void method() { ... }
} 

package c;
public class C extends A {
  void method() { ... }
}

When using Resolver#overrides(methodC, methodA) we get different results in KSP1 and KSP2:

• KSP1: Resolver#overrides(methodC, methodA) == false
• KSP2: Resolver#overrides(methodC, methodA) == true

I think KSP2's result is wrong in this case because A#method() is not visible from class B, so rather than an override it's more like shadowing.

[bcorso]

The jvm spec seems to also suggest this (https://docs.oracle.com/javase/specs/jvms/se8/html/jvms-5.html#jvms-5.4.5):

An instance method mC declared in class C overrides another instance method mA declared in class A iff either mC is the same as mA, or all of the following are true:

• C is a subclass of A.
• mC has the same name and descriptor as mA.
• mC is not marked ACC_PRIVATE.
• One of the following is true:
  • mA is marked ACC_PUBLIC; or is marked ACC_PROTECTED; or is marked neither ACC_PUBLIC nor ACC_PROTECTED nor ACC_PRIVATE and A belongs to the same run-time package as C.
  • mC overrides a method m' (m' distinct from mC and mA) such that m' overrides mA.

In particular, the bolded bullet point above is false in the example mentioned in #2254 (comment), so I think Resolver#overrides(methodC, methodA) should also give false.

[bcorso]

(I also just realized I filed something very similar for KSP1 #1279)