Changes proposed in this Pull Request

Good day. I was working with PyPSA and noticed that get_clustering_from_busmap function that are used in PyPSA-Eur and PyPSA-Earth makes the clustered network have timeseries for storage units filled with default. As a result, it was not possible to use aggregateoneport function again after clustering, because timeseries were not empty. Currently, get_clustering_from_busmap do not import empty timeseries for generators, links, and lines. However, for one port components like storage_units we have empty timeseries added with _import_series_from_df method, which fills missing inputs with defaults. For p, it is 0. It might be a not a huge problem, because after clustering in PyPSA-X we have exporting to netcdf, which cleans up default timeseries data. But it seems not convenient if you would like to do further manipulations with clustered network.

Is it intentional? Or we can keep one port components consistent with others and include if not df.empty? What do you think @lkstrp?

Checklist

• Code changes are sufficiently documented; i.e. new functions contain docstrings and further explanations may be given in docs.
• Unit tests for new features were added (if applicable).
• A note for the release notes docs/release-notes.md of the upcoming release is included.
• I consent to the release of this PR's code under the MIT license.