Add additional CI tests: Mutual Information, Chi square, Monte Carlo

[adam2392]

LGTM. My only issues are with some of the type hints.

@adam2392 can you comment here on why some flavors are left out? For example:

• mutual information and asymptotic chi-square test for discrete variables, perhaps with a James-Stein estimator
• Monte Carlo permutation testing and sequential Monte Carlo permutation testing
• Jonckheere-Terpstra test for ordinal variables

Originally posted by @robertness in https://github.com/py-why/dodiscover/pull/16#pullrequestreview-1076586407

[adam2392]

Other tests that I'm trying to use for some projects are:

• [x] CCIT: https://arxiv.org/pdf/1709.06138.pdf - #28
• [x] CMI: https://proceedings.mlr.press/v84/runge18a/runge18a.pdf - #83
• [ ] CCMI: https://arxiv.org/abs/1906.01824 - #85
• [ ] Monte Carlo utilities - #85

[adam2392]

Additional info on the Jonckheree test:

https://www.quantics.co.uk/blog/jonckheere-terpstra-test-ordered-differences/ https://www.stats.ox.ac.uk/~steffen/teaching/fsmHT07/fsm07.pdf https://en.wikipedia.org/wiki/Jonckheere%27s_trend_test

C implementation in R: https://github.com/cran/bnlearn/blob/master/src/tests/discrete/jonckheere.c