MINOR: [Python] improve error log to return a list of duplicated columns

[adrienpacifico]

Improve error log to display the list of duplicated columns instead of the full columns list.

Rationale for this change

Getting the full list of columns of the dataframe is not usefull, knowing which columns is duplicated is.
I tought that having all duplicated columns of the dataframe is better than having only the set of duplicated column names (df.columns[df.columns.duplicated()].unique().tolist() )

Are these changes tested?

No, very minor changes. Checked that f-strings are compatible with Python 3.9 (oldest compatible Python version in the pyproject.toml .

Are there any user-facing changes?

No

[github-actions]

Thanks for opening a pull request!

If this is not a minor PR. Could you open an issue for this pull request on GitHub? https://github.com/apache/arrow/issues/new/choose

Opening GitHub issues ahead of time contributes to the Openness of the Apache Arrow project.

Then could you also rename the pull request title in the following format?

GH-${GITHUB_ISSUE_ID}: [${COMPONENT}] ${SUMMARY}

or

MINOR: [${COMPONENT}] ${SUMMARY}

See also:

• Other pull requests
• Contribution Guidelines - Contributing Overview

[AlenkaF]

Hi @adrienpacifico, thank you for the suggested contribution! I would very much like if the test test_duplicate_column_names_does_not_crash in test_pandas is updated to reflect this change and also an issue opened so this, even if small diff, can be tracked.