FIX: Add message for single column in associations tab

[mithilP007]

Fixes #1747

Adds a conditional check to display a clear message when there is only one column in the dataframe, instead of showing the misleading "No strong associations" message.

The new message: "No associations computed because there is only one column."

[mithilP007]

The test failure on test (macos-latest, ci-py309-min-deps) is due to a CI infrastructure issue, not related to the code changes in this PR.

The error shows a network timeout while fetching dependencies:

× failed to solve the pypi requirements of 'ci-nightly-deps'
└─► operation timed out

This is a transient issue with accessing https://pypi.anaconda.org/scientific-python-nightly-wheels/simple/pyarrow/. All other 23 checks passed successfully. The code changes are working as intended.

[rcap107]

The test failure on test (macos-latest, ci-py309-min-deps) is due to a CI infrastructure issue, not related to the code changes in this PR.

The error shows a network timeout while fetching dependencies:

× failed to solve the pypi requirements of 'ci-nightly-deps'
└─► operation timed out

This is a transient issue with accessing https://pypi.anaconda.org/scientific-python-nightly-wheels/simple/pyarrow/. All other 23 checks passed successfully. The code changes are working as intended.

I restarted the jobs, everything seems to be working now

[rcap107]

Hi @mithilP007, thanks for the PR. Could you please add a small test in test_table_report.py that checks that the correct message is shown when the associations are skipped? Both when they are skipped because the number of columns is larger than max_association_columns, and when there is only one column (this specific PR).

We can merge it after the test is added.

[MarieSacksick]

Hi @mithilP007 :)!
Could you please remove the changes about # noqa: F401 in scikit-learn compat file? It looks like it's breaking the CI, plus it's not related to the goal fo the PR.

You can do pre-commit run --all-files locally to check that this part of the CI passes. (cf the contributing guide to know more about it).

[rcap107]

Thanks for the comment @MarieSacksick, you're completely right. I must have lost track of this PR.

@mithilP007 as Marie mentioned, the latest commit is not useful for the PR, so it should be reverted.

Do you need help with the test?