Standardize dataset information for SEVIRI and FCI L2 products

[strandgren]

The dataset information provided in the reader yaml-files for the SEVIRI and FCI L2 products should be standardized following the CF conventions where possible making use of standard_name and long_name in a consistent manner. For categorical data the CF Flags (flag_values, flag_meanings) shall be added as well.

See documentation here: http://cfconventions.org/Data/cf-conventions/cf-conventions-1.9/cf-conventions.html#_description_of_the_data

Please assign this issue to me.

[strandgren]

Following the merge of the two PRs #2341 and #2665, the SEVIRI and FCI L2 readers have now been modified in order for the dataset to follow the CF convention where possible and make sure that standard_name, long_name and units are available as dataset attributes and populated properly. Furthermore, flag_values and flag_meanings attributes have been added for all categorical datasets. Hence, this issue can now be closed.