Aquaplanet rerun with current version

[MarionBWeinzierl]

We should do an aquaplanet rerun with the current version (after issues with mixing ratios/number concentration tendencies fixed) and check if striped bands present and/or due to aspect of implementation.

This is related to #41 since this could be a source of the banding. Another thing to look at would be array sizes and a mismatch between pcols , ncols, and the array size used by the parameterisation.

[MarionBWeinzierl]

#41 has been resolved #40 with #50, including an aquaplanet rerun.

We are continuously rerunning the aquaplanet with the latest changes.