TST/CLN: fix warning "gpu available but not used"

[NickleDave]

When I run tests on a machine with a GPU I get a warning from lightning that we're not using gpu even though it's detected

tests/test_eval/test_frame_classification.py: 5 warnings
tests/test_eval/test_parametric_umap.py: 1 warning
tests/test_learncurve/test_frame_classification.py: 8 warnings
tests/test_predict/test_frame_classification.py: 2 warnings
tests/test_train/test_frame_classification.py: 4 warnings
tests/test_train/test_parametric_umap.py: 1 warning
  /home/ildefonso/Documents/repos/vocalpy/vak-vocalpy/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/setup.py:187: GPU available but not used. You can set it by doing `Trainer(accelerator='gpu')`.

• figure out if this is because of how we configure things in tests or because of how the code actually runs
• if the latter, can we do something to squelch this warning? e.g. directly specify acclerator='cpu' so that it's clear our intent is not to use GP