Preserve schema during column re-naming in xgboost and lightgbm

[freddyaboulton]

Pull Request Description

The fact that we don't preserve the schema during column re-naming at the last step of xgboost + lightgbm pipelines makes it so that this unrelated ww bug (alteryx/woodwork#1411) causes some of our pipelines to fail in corner cases.

This is a fix to prevent pipelines from crashing. We should have been preserving the schema anyways.

Perf tests

report.html.zip

---

After creating the pull request: in order to pass the release_notes_updated check you will need to update the "Future Release" section of docs/source/release_notes.rst to include this pull request by adding :pr:123.

[codecov]

Codecov Report

Merging #3496 (5475c56) into main (3251a29) will decrease coverage by 0.1%.
The diff coverage is 100.0%.

@@           Coverage Diff           @@
##            main   #3496     +/-   ##
=======================================
- Coverage   99.7%   99.7%   -0.0%     
=======================================
  Files        335     335             
  Lines      33178   33238     +60     
=======================================
+ Hits       33050   33109     +59     
- Misses       128     129      +1     

Impacted Files	Coverage Δ
...nents/estimators/classifiers/xgboost_classifier.py	100.0% <100.0%> (ø)
...ponents/estimators/regressors/xgboost_regressor.py	100.0% <100.0%> (ø)
...alml/tests/component_tests/test_lgbm_classifier.py	100.0% <100.0%> (ø)
...valml/tests/component_tests/test_lgbm_regressor.py	100.0% <100.0%> (ø)
...l/tests/component_tests/test_xgboost_classifier.py	100.0% <100.0%> (ø)
...ml/tests/component_tests/test_xgboost_regressor.py	100.0% <100.0%> (ø)
evalml/tests/utils_tests/test_gen_utils.py	100.0% <100.0%> (ø)
evalml/utils/gen_utils.py	99.2% <100.0%> (-0.3%)	⬇️

---

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 3251a29...5475c56. Read the comment docs.

[eccabay]

LGTM! Just left a few small comments.

[eccabay]

Nit: classifier should probably be named lightgbm not xgb here

[eccabay]

Same naming note here

[eccabay]

If I'm reading this correctly, it seems like we no longer ever set flatten_tuples to False. In that case, are we ok to just remove the argument entirely?

[chukarsten]

Agreed with Becca's comments about variable names and flatten_tuples! Thank you!