Support for Multiclass Classification in Class Imbalance data checks

[bchen1116]

fix #1864

Summary:

• Compute the number of samples in each class of the training data
• Identify the majority class which has the largest number of samples
• For each class, compute the ratio of majority class samples to that classes sample count
• If that ratio exceeds a threshold (default 9:1), raise a class imbalance warning
• If that ratio exceeds a threshold AND the number of samples for that class < 100, raise a severe class imbalance warning

Did not incorporate steps for number of classes in order to keep this data check a little simpler and keep the logic in line with what users would expect. (ie at larger numbers of classes, we don't automatically lower the threshold, even though it might be imbalanced for the model to learn).

[codecov]

Codecov Report

Merging #1986 (73a7934) into main (2afdc84) will increase coverage by 0.1%.
The diff coverage is 100.0%.

@@            Coverage Diff            @@
##             main    #1986     +/-   ##
=========================================
+ Coverage   100.0%   100.0%   +0.1%     
=========================================
  Files         274      274             
  Lines       22286    22299     +13     
=========================================
+ Hits        22280    22293     +13     
  Misses          6        6             

Impacted Files	Coverage Δ
evalml/data_checks/class_imbalance_data_check.py	100.0% <100.0%> (ø)
...ta_checks_tests/test_class_imbalance_data_check.py	100.0% <100.0%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2afdc84...73a7934. Read the comment docs.

[dsherry]

Looks great! I left a couple impl comments

[angela97lin]

Neat, I'm a big fan of this improvement! LGTM 🥳

[ParthivNaresh]

Very cleanly done, I love it