Reading Series with columns=[] doesn't return the index

[G-D-Petrov]

Currently, reading Series with columns=[] returns an empty Series, instead of Series with the index like for DataFrame. The is because Pandas doesn't support Series with index but empty data. The use case is illustrated with this test:

def test_series_read_column_empty(lmdb_library_static_dynamic, sym):
    lib = lmdb_library_static_dynamic
    series = pd.Series(np.arange(366), name="test", index=pd.date_range("2020-01-01", "2020-12-31"))
    lib.write(sym, series)
    res = lib.read(sym, columns=[]).data
    assert res.empty
    assert res.index.equals(series.index)