(🎁) support docstrings on `type` statements about cpython

[sobolevn]

I think that this would be really uncommon for a non-block statement to support docstrings this way.

Right now __doc__ is not even assignable on TypeAliasType instances:

>>> type A = int
>>> A.__doc__
'Type alias.\n\nType aliases are created through the type statement::\n\n    type Alias = int\n\nIn this example, Alias and int will be treated equivalently by static\ntype checkers.\n\nAt runtime, Alias is an instance of TypeAliasType. The __name__\nattribute holds the name of the type alias. The value of the type alias\nis stored in the __value__ attribute. It is evaluated lazily, so the\nvalue is computed only if the attribute is accessed.\n\nType aliases can also be generic::\n\n    type ListOrSet[T] = list[T] | set[T]\n\nIn this case, the type parameters of the alias are stored in the\n__type_params__ attribute.\n\nSee PEP 695 for more information.\n'
>>> A.__doc__ = 'My info'
Traceback (most recent call last):
  File "<python-input-2>", line 1, in <module>
    A.__doc__ = 'My info'
    ^^^^^^^^^
AttributeError: 'typing.TypeAliasType' object attribute '__doc__' is read-only

[AlexWaygood]

I don't think it's unreasonable to want to add a docstring to a type alias, but I agree with @sobolevn that this specific proposal would be pretty unprecedented. So we'd need to have a discussion over exactly how we'd want to add the feature. In short, I agree with @nineteendo — this would need to first be discussed on Discourse to hash out design details of the feature in a community forum, before opening a GitHub issue

[nineteendo]

Major feature proposals should generally be discussed on Discourse before opening a GitHub issue.