[com4FlowPy] Implement "Back-Tracking"/"Back-Calculation" feature more efficiently

[ahuber-bfw]

The "back-calculation" or "back-tracking" function in com4FlowPy, which is also described in D'Amboise et al. (2022) --> https://gmd.copernicus.org/articles/15/2423/2022/ and discussed in D'Amboise et al. (2021) --> https://www.intechopen.com/chapters/78979 is currently implemented in flowCore.py.

This is a super cool/important feature, but currently suffers from inefficient code-implementation (Runtimes with provided infrastructure-layer / "back-tracking" enabled take a lot longer, than model runs without this feature.

• think about a smarter/more efficient way to implement the "back-tracking" feature in flowCore.py
• Desirable would be to have model runtimes with "back-calculation"/"infrastructure" max. twice as long as model runtimes without the feature (now runtimes with back-tracking are heavily dependend on the number of infrastructure cells, that get "hit"/affected by modeled processes and in test-runs it is not uncommon, that model runs with back-tracking feature enabled are taking 15x as long to process)