BUG: pd.unique() does not accept NumpyExtensionArray

[mutricyl]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd
arr_int = pd.array([1, 2, 3]) # Int64Dtype() => ExtensionArray
arr_complex = pd.array([1+1j, 2, 3]) # NumpyEADtype('complex128') => NumpyExtensionArray

pd.unique(arr_int)  # OK
pd.unique(arr_complex)  # NOK: TypeError: unique requires a Series, Index, ExtensionArray, or np.ndarray, got NumpyExtensionArray.

Issue Description

This issue is similar to #59177 and is coming from pint-pandas where test CI job with nightly builds from pandas was recently introduced (pint-pandas issue related to skipped tests to pass tests with pandas 3).

The above example is fine with pandas 2.2.2 but fails with pandas nightlies

The above mentionned TypeError is coming from

pandas/pandas/core/algorithms.py

Line 222 in ab433af

def _ensure_arraylike(values, func_name: str) -> ArrayLike:

As explained in #59177 NumpyExtensionArray may be used internally when other type of ExtensionArray are not available.

Expected Behavior

for pandas 2.2.2 pd.unique returns:

>>> pd.unique(arr_complex)
array([1.+1.j, 2.+0.j, 3.+0.j])

While we expect

>>> pd.unique(arr_complex)
<NumpyExtensionArray>
[(1+1j), (2+0j), (3+0j)]
Length: 3, dtype: complex128

Installed Versions

Replace this line with the output of pd.show_versions()

[ritwizsinha]

take

[ritwizsinha]

Added a similar PR like #59181