dataclass overriding a property in the parent class is broken

Bug report

Bug description:

Consider the following snippet

@dc.dataclass
class A:
  a: int
  @property
  def b(self) -> int:
    return 1

@dc.dataclass
class B(A):  
  b: int = dc.field(default_factory=lambda: 0)

Dataclass apparently doesn't create a class attribute, so class B ends up with property , b of class A, which seems surprising, especially given that there is explicit assignent of field b in class B(A).

So this results in:

>>> B.b
<property object at ...>
>>> B(a=2)
Traceback (most recent call last):                                                                                                                            
  File "<stdin>", line 1, in <module>                                                                                                                         
  File "<string>", line 4, in __init__                                                                                                                        
AttributeError: property 'b' of 'B' object has no setter      

If i replace definition of b like so b: int = dc.field(default=0), or with b: int = 0 then everything works as expected:

>>> B.b
0
>>> B(a=2)
B(a=2, b=0)

I think it would be nice if dataclass would always create a class attribute (especially if there is explicit assignment!)
so that overloads work as expected.

CPython versions tested on:

3.11, 3.12

Operating systems tested on:

Linux