Robustify post process

[malihass]

• Some fields where expected for instantaneous kla calculation and were not always there. We now force these fields to be present in the tutorial and experimental cases
• d.gas is not present if we run a constant bubble size diameter case. If it is absent, we read d.gas from phaseProperties.
• If thermo:mu.liquid is not present, we check that we are using a constant viscosity model and read the muMixLiq from globalVars
• Computation of kla did not handle passing the species as str while it should be allowed. If passed as string, species are internally converted to list of 1 species.
• Compute KLA with the fitting method : separate kla calculation from the file reading so that the same function can be used for the tutorial and post_quantities. Change the logging to show the units.
• Read species molecular weight from the thermophysical properties first rather than globalVars. Global vars is the last resort. Add a test for this workflow