One Hot Encoder: Drop one redundant feature by default for features with two categories

[angela97lin]

Closes #1936

Logic for our own custom dropping for binary features (features with only two categories):

1. During fit, if is_binary, determine which features have two categories and which is the majority feature which should be dropped. Store this information. Pass None to scikit-learn OHE so that scikit-learn does not drop.
2. Transform. Get output array from scikit-learn. During get_feature_names, find the name of the column that we should drop. Ex: col "original" with majority category "a" might have a transformed name of "original_a"; store "original_a" so we know to drop this column. Before returning, drop all columns that have been specified.

---

Impl notes:

• separates out _get_feature_names from get_feature_names, where _get_feature_names is pre-dropping binary features and more for our private implementation, while get_feature_names is user-facing and has the expected cols (without the binary feature that is dropped)

[codecov]

Codecov Report

Merging #1997 (b199d28) into main (2cbfa34) will increase coverage by 0.1%.
The diff coverage is 100.0%.

@@            Coverage Diff            @@
##             main    #1997     +/-   ##
=========================================
+ Coverage   100.0%   100.0%   +0.1%     
=========================================
  Files         274      274             
  Lines       22325    22360     +35     
=========================================
+ Hits        22319    22354     +35     
  Misses          6        6             

Impacted Files	Coverage Δ
evalml/tests/component_tests/test_components.py	100.0% <ø> (ø)
...components/transformers/encoders/onehot_encoder.py	100.0% <100.0%> (ø)
...alml/tests/component_tests/test_one_hot_encoder.py	100.0% <100.0%> (ø)
evalml/tests/pipeline_tests/test_pipelines.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 2cbfa34...b199d28. Read the comment docs.

[freddyaboulton]

@angela97lin Nice! This was a surprisingly tricky one. I think this looks good. I have some minor non-blocking comments and a question on test_ohe_column_names_unique

[bchen1116]

Nice! LGTM