SI3: optimization and debug of salinity drainage and flushing

Salinity drainage and flushing in the ice has been implemented a couple of months ago following a paper from Thomas et al. 2020. However, in the context of an internship, we shed some light on deficiencies of this implementation. In particular, the scheme from ReeseJones and Worster was bugged. In addition, the routine was very consuming in terms of cpu. This issue is raised to correct these problems and to tune the parameterization, so that it fits the observations.