Expand data validation to also cover the scikit-learn experiment classes

[drbenvincent]

• Closes #78
• We currently have _input_validation methods in just the pymc experiment classes
• This PR extracts these out into new Mixin classes in data_validation.py
• Now we can use the same data validation code for both the pymc and the scikit-learn versions of the experiment classes.

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 97.64706% with 2 lines in your changes are missing coverage. Please review.

Project coverage is 77.06%. Comparing base (67d9e0e) to head (6574481).

Files	Patch %	Lines
causalpy/data_validation.py	96.15%	2 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #290 +/- ## ========================================== + Coverage 76.50% 77.06% +0.56% ========================================== Files 20 21 +1 Lines 1345 1378 +33 ========================================== + Hits 1029 1062 +33 Misses 316 316 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[drbenvincent]

Thanks @jpreszler. I've added tests for the scikit-learn versions of the experiments in test_input_validation.py