An abstract base class for the data backends: DoubleMLBaseData

[MalteKurz]

Description

This PR introduces a new abstract base class DoubleMLBaseData The goal is to disentangle the basic data handling, from the handling of the causal model (i.e. which variable is the treatment variable etc.).

Comments

The overall goal is to make it easier to write new data backends. Note however that the handling of the multiple treatment case is currently a pretty fundamental / core part of the implementation of the base class DoubleML. Therefore, it wasn't yet possible to completely get rid of properties like d_cols or n_treats in the base class DoubleMLBaseData.

PR Checklist

• The title of the pull request summarizes the changes made.
• The PR contains a detailed description of all changes and additions.
• The code passes all (unit) tests.
• Enhancements or new feature are equipped with unit tests.
• The changes adhere to the PEP8 standards.