return molar density from fluidstate in volumevariables. this is a more consistent implementation of molar density for co2
The old implementation was inconsistent because when using the policy: useCO2GasDensityAsGasMixtureDensity the density was calculated as the mixture of the gases while the molar density was calculated as the co2 molar density only and then divided by the average molar mass in the volumevariables. This does not affect the co2 tests on dumux- master but in the lecture it changes the result for the co2 plume