Handle warnings output by tests

[riley-harper]

There are four unique warnings output when the tests run (they are repeated many times though). Three of the warnings are deprecation warnings coming from PySpark, which we don't have much control over. Hopefully they will be fixed in future versions of PySpark; we'll see.

The last warning is

/usr/local/lib/python3.10/site-packages/sklearn/metrics/_ranking.py:874: UserWarning: No positive class found in y_true, recall is set to one for all thresholds.

This is caused by the fact that some of our test sets are very small, and so might not have a good mix of positive and negative classifications. This isn't important for this test, so let's try to ignore or silence the error in pytest. It would be good if we could scope this to the single test instead of doing it globally for all tests.