More Pre-Pocessing Functionality for Relations

It would be interesting to get more functioanlity to the `relations`, that would allow to read multiple results from the simulation and perfomr some (arbitrary) computation on them. Helpful if quantities can not directly be determined by the simulation software, which are to be used as optimisation target.

The issue does not name files, tests, or an entry point. Start by tracing how relations expose simulation results, then clarify how multiple results and arbitrary computations should be represented and consumed as optimization targets; done requires an agreed interface and documented behavior.

Written by the indexing model from the issue text.