feat: either compare across splits or crossvalidate

[sarakolding]

• [x] I have battle-tested on Overtaci (RMAPPS1279)
• [x] At least one of the commits is prefixed with either "fix:" or "feat:"

Notes for reviewers

Reviewers can skip X, but should pay attention to Y.

[github-actions[bot]]

✨ Looks like pyright failed ✨

If you want to fix this, we recommend doing it locally by either:

a) Enabling pyright in VSCode and going through the errors in the problems tab

VSCode settings > Python > Analysis: Type checking mode > "basic"

b) Debugging via the command line

1. Installing pyright, which is included in the dev dependencies: `pip install -e ".[dev]"`
2. Diagnosing the errors by running `pyright .`

[github-actions[bot]]

Test Results

23 tests   23 :heavy_check_mark:  4m 18s :stopwatch:   1 suites    0 :zzz:   1 files      0 :x:

Results for commit 092ca1ee.

:recycle: This comment has been updated with latest results.

[github-actions[bot]]

Tests	Skipped	Failures	Errors	Time
22	0 :zzz:	0 :x:	0 :fire:	2m 55s :stopwatch:

[github-actions[bot]]

🌟 pyright succeeds! 🌟

[sarakolding]

looking back, it's probs not necessary to add a synth test set to test the train_val_predict function, but it made sense in my brain at the time 🙃