Extended time series regularizer to support multiseries

[MichaelFu512]

Pull Request Description

Extended time series regularizer to support multiseries
Closes #4304

Example of what this does

Let's say you have an unstacked X:

Where "2018-01-11", "2018-01-12" and "2018-01-16" are missing.

And you have an unstacked y:

Which is missing the values that correspond to the missing dates from X_unstacked (values jumps from 49 -> 60 and 74 -> 80).

If you call TimeSeriesRegularizer on X and y:

y will now regularize into:

Works also for duplicate datetimes, extra datetimes, and misaligned datetimes.

---

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 (cc60b68) 99.7% compared to head (e90a7ea) 99.7%.

Additional details and impacted files

@@           Coverage Diff           @@
##            main   #4303     +/-   ##
=======================================
+ Coverage   99.7%   99.7%   +0.1%     
=======================================
  Files        357     357             
  Lines      39694   39739     +45     
=======================================
+ Hits       39574   39619     +45     
  Misses       120     120             

Files Changed	Coverage Δ
...nsformers/preprocessing/time_series_regularizer.py	100.0% <100.0%> (ø)
...ts/component_tests/test_time_series_regularizer.py	100.0% <100.0%> (ø)

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

[eccabay]

We're going to need more testing, but the impl itself is solid!

[jeremyliweishih]

Quick refactor but otherwise LGTM.

[eccabay]

A few small comments, but nothing blocking. The tests are much cleaner now, thank you!