Multiple issues with StrInCol #18

Reproduce:

df = pd.read_csv('./datasets/dataranch__supermarket-sales-prediction-xgboost-fastai__SampleSuperstore.csv')
if '41' in df['Profit'].to_string():
    pass

This errors, since somewhere in the rewriting process, df['Profit'] becomes df[Profit], which causes NameError: name 'Profit' is not defined.

df_multi_index = df.set_index(['Postal Code', 'Sub-Category'])
for col in df_multi_index.columns:
    if 'ou' in df_multi_index[col].to_string():
        pass

This also errors, since the existing rewriting rule (which uses index.astype(str)) doesn't work for a MultiIndex, yielding TypeError: Setting a MultiIndex dtype to anything other than object is not supported.