Added clarification on self-dependencies in model structure

[PeterMeisrimelModelon]

Fixes #2055

Might need some polishing on the exact formulation, but should be a good-enough starting point.

[chrbertsch]

FMI Design Meeting:

Peter: in the best case also the self-dependency should be listed. The issue came up on a reference FMU.
Pierre: It is just stated that a variable does not have list the dependency on itself. This occures since FMI 3.0 as states are exposed as normal variables.