allow template_geographic_coverage to take a vector of tables rather than just one

[RobLBaker]

Currently, template_geographic_coverage requires that geographic data be submitted as a single file. But, EML may describe multiple different files each with it's own geographic coordinates.

Rather than using a join statement to make an additional file that contains all the geographic coordinates from all the datasets, would it be possible for template_geographic_coverage to take a vector or list of data tables as input, much as template_taxanomic_coverage does?

[clnsmth]

Thanks @RobLBaker. This request and proposed pattern make a lot of sense, especially in light of the template_taxonomic_coverage() implementation.

One thing an updated template_geographic_coverage() could (should?) do is handle duplicate geographic coverages gathered from the multiple input tables, where duplicates are non-unique combinations of site name, latitude, and longitude. These could be dropped with a warning that lists the dropped information. Would something like this work?

[RobLBaker]

That sounds great! Adding the site name to the non-unique combination heads off some potential problems with people naming the same geographic coordinates different things (e.g. if "Veg_site_1" is at the same location as "Vert_site_1")