feat: support inplace=True in rename and rename_axis

[tswast]

Thank you for opening a Pull Request! Before submitting your PR, there are a few things you can do to make sure it goes smoothly:

• Make sure to open an issue as a bug/issue before writing your code! That way we can discuss the change, evaluate designs, and agree on the general idea
• Ensure the tests and linter pass
• Code coverage does not decrease (if any source code was changed)
• Appropriate docs were updated (if necessary)

Fixes internal issue b/329460931 🦕

[tswast]

• Missed Index.rename. Will mark as ready once that's up.

Edit. Done!

[tswast]

Re: failures

____________________________ test_loc_list_nameless ____________________________ [gw2] linux -- Python 3.9.20 /tmpfs/src/github/python-bigquery-dataframes/.nox/system-3-9/bin/python scalars_df_index = bool_col bytes_col \ rowindex ....999999+00:00 8 T <NA> <NA> [9 rows x 13 columns] scalars_pandas_df_index = bool_col ... timestamp_col rowindex ... 0 ... ... 2038-01-19 03:14:17.999999+00:00 8 False ... <NA> [9 rows x 13 columns] def test_loc_list_nameless(scalars_df_index, scalars_pandas_df_index): index_list = [0, 0, 0, 5, 4, 7] bf_series = scalars_df_index.string_col.rename(None) bf_result = bf_series.loc[index_list] pd_series = scalars_pandas_df_index.string_col.rename(None) pd_result = pd_series.loc[index_list] > pd.testing.assert_series_equal( bf_result.to_pandas(), pd_result, ) E AssertionError: Series.index are different E E Attribute "names" are different E [left]: [None] E [right]: ['rowindex'] 

I think I need to call .copy() before I do anything inplace.

Edit: this hypothesis seems to be incorrect. I don't add any system tests for inplace, only unit tests. Instead, it was the problem identified with tuples.

[tswast]

I think this change to Label (Hashable) is what's breaking some MultiIndex tests. Need a better way to check if something is already a sequence / iterable and not a string.

[tswast]

Fixed in 45f9232

I think we want to support integer names, too, so instead I specifically exclude tuple from the check for hashable objects.

[tswast]

e2e failures:

FAILED tests/system/small/ml/test_linear_model.py::test_linear_reg_model_global_explain FAILED tests/system/small/ml/test_preprocessing.py::test_label_encoder_series_default_params FAILED tests/system/small/ml/test_preprocessing.py::test_one_hot_encoder_save_load FAILED tests/system/small/ml/test_preprocessing.py::test_standard_scaler_normalizes FAILED tests/system/small/ml/test_preprocessing.py::test_label_encoder_params FAILED tests/system/small/ml/test_preprocessing.py::test_standard_scaler_normalizeds_fit_transform FAILED tests/system/small/ml/test_llm.py::test_llm_gemini_pro_score_params[gemini-1.5-pro-002] 

All of these seem to be timeout related with BQML integration. I don't think it should have been caused by my changes.

Presubmits are still running, but the 3.9 system tests passed.