Addition of a new wind speed-dependent dendricity and sphericity paramterization

This MR includes 2 changes:

1. Addition of a new wind speed-dependent dendricity and sphericity paramterization, combining elements from CROCUS' (Vionnet 2012) and Zwart's 2007 parameterizations. For operational use in Norway, we have found improved representation of rounded grains with this parameterization (in conjunction with the hn_density_parameterization VANKAMPENHOUT), whereas otherwise we found an overrepresentation of facets in the upper layers. This is likely due to the fact that with any hn_density_parameterization that is not LEHNING_NEW, ZWART, JORDY or BELLAIRE, the sphericity & dendricity of new snow are 0.5 and 1.0 respectively, regardless of wind speed. (See solid red lines in plots below).

For now, this new parameterization is selected if none of the other clauses are met (i.e. hn_density_param is not LEHNING_NEW, ZWART, JORDY or BELLAIRE) AND vw_dendricity is true (which it is, hardcoded). This does however raise the question if we want these dendricity and sphericity parameterization(s) dependent on a combination of hn_density_parameterization, vw_dendricity, and Variant or if we should introduce a separate hn_microstructure_parameterization? These multi-clause if-statements become increasingly complex to read as we keep adding to them.

2. Restructuring of the if-statements for new snow dendricity, sphericity, bond size and grain size. This should make the code more understandable, but does not change its function. Specifically:

• Lead with the default case for sp & dd, instead of the least used option (Jordy).
• separate blocks for dd&sp, and rg&rb.