Extend TimeSeriesImputer to handle multiple series

[MichaelFu512]

Pull Request Description

Extend TimeSeriesImputer to handle multiple series

Closes #4292

Example picture of what this does

This is the unstacked y from an example multiseries timeseries problem.

The imputer now imputes on it:

---

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

Patch coverage: 100.0% and project coverage change: +0.1% 🎉

Comparison is base (69344b2) 99.7% compared to head (41c277e) 99.7%.

Additional details and impacted files

@@           Coverage Diff           @@
##            main   #4291     +/-   ##
=======================================
+ Coverage   99.7%   99.7%   +0.1%     
=======================================
  Files        355     355             
  Lines      39458   39511     +53     
=======================================
+ Hits       39338   39391     +53     
  Misses       120     120             

Files Changed	Coverage Δ
evalml/pipelines/components/component_base.py	100.0% <ø> (ø)
...nents/transformers/imputers/time_series_imputer.py	100.0% <100.0%> (ø)
.../tests/component_tests/test_time_series_imputer.py	100.0% <100.0%> (ø)
...alml/tests/utils_tests/test_nullable_type_utils.py	100.0% <100.0%> (ø)
evalml/utils/nullable_type_utils.py	100.0% <100.0%> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[eccabay]

Just some small changes, but this looks great!

[eccabay]

I think we would benefit from another test (parametrized here) where only some of the columns are NaN, but not all of them!

[eccabay]

^ Sorry, should have clarified 😅 I meant a test where some of the columns are all NaN, so we drop some columns and impute or pass through others!

[eccabay]

Just a few more comments!

[eccabay]

^ Sorry, should have clarified 😅 I meant a test where some of the columns are all NaN, so we drop some columns and impute or pass through others!

[eccabay]

Thanks!

[jeremyliweishih]

LGTM!