[dask] Add unit tests that signatures are the same between Dask and scikit-learn estimators

[ghost]

I tried to check for a sublist in sklearn's args which matches dask's (this ensures that order is checked), the same goes for the
default values.

fixes #3907

[ghost]

All CLA requirements met.

[ghost]

bunch of syntax and styling mistakes, my bad

[jameslamb]

Thanks for picking this up! I like the way you set this up with pytest.mark.parametrize. I left some suggestions for your consideration.

[jameslamb]

I've edited your pull request description to include fixes #3907. This creates a link to that issue and will automatically close that issue when the PR is merged.

[jameslamb]

Thanks! Please see my newest comments.

[ghost]

Now I understand why I was getting the TypeError: object of type 'NoneType' has no len().
It was because of that same issue that sklearn's fit() methods don't have **kwargs pass-through.

[jameslamb]

thanks! Please see my newest suggestion

[jameslamb]

this looks great, thanks very much for the contribution!

[ghost]

Thanks a lot for the guidance!