Update Var aggregation

[SvenKlaassen]

Description

Change variance aggregation to scaled level (scaled variances are aggregated instead of applying the rescaling afterwards).
Enable the option to apply different scalings at different repretitions in the utilities (not used in the current implementation).

Comments

The different scaling are not necessary in the DoubleML class since all the scalings are supposed to be of the same size.
This might be relevant for future RDD implmentations.

PR Checklist

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