BUG: PandasArray.to_numpy mishandles na_value

[mzeitlin11]

Problem is here:

pandas/pandas/core/arrays/numpy_.py

Lines 358 to 364 in acacff3

	result = np.asarray(self._ndarray, dtype=dtype)
	if (copy or na_value is not lib.no_default) and result is self._ndarray:
	result = result.copy()
	if na_value is not lib.no_default:
	result[self.isna()] = na_value

Since na_value is handled after the dtype conversion, calls like

arr = pd.array([pd.NA, 1], dtype="string")
arr.to_numpy(na_value=True, dtype=bool)

fail.

[Sudomarko]

take