Fix oversampler data typing error

[eccabay]

Provides a workaround for an issue where categorical datatypes were causing the SMOTENC oversampler to fail. The SMOTENC oversampler can only handle object datatypes and not categorical. This was manifesting itself in CatBoost pipelines, since we don't one-hot encode the categorical features for CatBoost.

[codecov]

Codecov Report

Merging #3732 (20b197d) into main (ce58f45) will increase coverage by 0.1%.
The diff coverage is 100.0%.

@@           Coverage Diff           @@
##            main   #3732     +/-   ##
=======================================
+ Coverage   99.7%   99.7%   +0.1%     
=======================================
  Files        338     338             
  Lines      34649   34659     +10     
=======================================
+ Hits       34517   34527     +10     
  Misses       132     132             

Impacted Files	Coverage Δ
...s/components/transformers/samplers/base_sampler.py	83.9% <100.0%> (+1.5%)	⬆️
evalml/tests/component_tests/test_oversampler.py	100.0% <100.0%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.