Pass the chemistry object back to the hydro chemistry diffusion interactions.

We'll have to think about how to properly initialise the object in the unit tests, however.

Thank you!

Ah I see. Are these unit tests subrid-model independent? This is a part of the code I have never looked at yet.

Ah the chemistry_data cannot be called chemistry data because of grackle module:

runner_doiact_functions_hydro.h: In function 'runner_dopair1_naive_force':
runner_doiact_functions_hydro.h:50:39: error: declaration of 'chemistry_data' shadows a global declaration [-Werror=shadow]
   50 |   const struct chemistry_global_data *chemistry_data = e->chemistry;
      |                                       ^~~~~~~~~~~~~~
In file included from /home/darwinr/local/include/grackle.h:18,
                 from ././cooling/grackle/cooling_properties.h:26,
                 from ././cooling/grackle/cooling.h:28,
                 from ./cooling.h:47,
                 from ./sink/GEAR/sink.h:33,
                 from ./sink/GEAR/sink_iact.h:26,
                 from sink_iact.h:31,
                 from runner_doiact_hydro.c:36:
/home/darwinr/local/include/grackle_chemistry_data.h:176:3: note: shadowed declaration is here
  176 | } chemistry_data;
      |   ^~~~~~~~~~~~~~

That's probably why I called it chem_data in the merge request.