Do not pass the chemistry model object to the diffusion loop. Extend the...

@matthieu Can I ask why you removed the chemistry data from the diffusion loop?

I introduced it two weeks ago for the GEAR FVPM diffusion scheme

I was cleaning things and saw it was unused in all schemes. But I should have checked whether that was indeed the case. I am not a fan of passing in so many properties things through the loops though but if we can't do it in another way then let's go ahead.

The FVPM diffusion is still not in master... Since the FVPM diffusion is a large branch with two submodules, I am merging it bit by bit. Two weeks ago, we merged the high-level API changes that touch everything outside the chemistry, including passing the chemistry properties to the loops.

I know... But in this case, I need the parameters for the Riemann solver and the diffusion flux stored in the chemistry props. We can discuss this once I merge this part for the parabolic diffusion.

The kiara team also needs it so I will re-add it.