Rewrite docstrings to match scikit-learn Style

We should try and match the way that scikit-learn writes their docstrings. [Here](https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/linear_model/_huber.py#L125) is an example.

For example, there is a blank line between the parameters and the default values are indicated right after the parameter type. It's also nice how the types that are numpy arrays also include the shapes.