reportRedeclaration diagnostics for redeclared property setters and deleters

[InSyncWithFoo]

Reproducible example (playground):

class C:

	def m(self) -> str: ...           # Method declaration "m" is obscured by a declaration of the same name
	def m(self) -> str: ...

	@property
	def p(self) -> int: ...           # Method declaration "p" is obscured by a declaration of the same name
	@property
	def p(self) -> int: ...

	@p.setter
	def p(self, v: int) -> None: ...  # No errors
	@p.setter
	def p(self, v: str) -> None: ...

	@p.deleter
	def p(self, v: int) -> None: ...  # No errors
	@p.deleter
	def p(self, v: str) -> None: ...

Since at runtime @p.setter just overwrites C.p with a new property object containing the latest setter method, I think it would make sense to consider this as a redeclaration and report it accordingly.

Instance attribute redeclarations with the same annotations also don't trigger this rule, but this seems to be intentional.

	a: int                            # Declaration "a" is obscured by a declaration of the same name
	a: str

	b: int                            # No errors
	b: int

[erictraut]

Pyright is working as intended here. While you're correct that @p.setter overwrites C.p with a new property object, this is such a common pattern that pyright (and other type checkers like mypy) special case it and do not treat it like a redeclaration.