`to_2d()` should seam all the data variables by default

[pmav99]

When you have more than one data variables, you need to call to_2d() for each one and merge the datasets afterwards.

It would be nice if, instead, you could just call:

ds_2d = to_2d(ds, grid=[xn, yn, tri3n])

If someone wants to filter the variables where the "seaming" happens, they should do it like this:

to_2d(ds[["elev", "depth"]], mesh=[...])

[brey]

I have researched that, however there are node valued variables that are not space variants for resample to work. Such is the case of e.g. wetdry_node variable which is boolean.

I have update the to_2d to get a list and perform the operation for all user-described variables with one call.