Make Glathida files download work in parallel

[JordiBolibar]

Right now the downloading of Glathida files with a pmap doesn't work. This should be fixed in the future in order to speed up the downloading of data for lots of glaciers.

[albangossard]

I'm not sure this issue is still a thing.

Currently in the code we use filter_missing_glaciers!(rgi_ids::Vector{String}, params::Parameters) (and not filter_missing_glaciers!(glaciers::Vector{Glacier2D}, params::Parameters) which by the way could be removed) and this downloads one CSV for all the glaciers. There is no need of parallelism.

Additionally, since glathida data are now preprocessed with Gungnir, the function get_glathida_path_and_IDs is no longer used and I think that it can be removed.

So to sum up we should remove get_glathida_path_and_IDs and one of the implementation of filter_missing_glaciers!. @JordiBolibar do you agree with this conclusion?

[JordiBolibar]

Yes, this is again legacy code. At the beginning we were doing a manual implementation of the downloading of Glathida. Later, this has been included officially in OGGM, so it's much easier.

[albangossard]

Thank for this precision! I took profit of #79 to remove the two aforementioned functions.