Differently from QuantumEpanetSimulator, the QuantumWNTRSimulator approach with vqls and qubo linear solvers does not converge properly, even with the simplest toy Net0.inp example. It looks like that the implemented WNTR algorithm results in very ill-posed A matrices.
If we want to also use QuantumWNTRSimulator, we should investigate this problem in depth, maybe in connection with #13.