Support EntitySchema identifiers as input to wdsub

[andrawaag]

I was able to reproduce run wdsub and would like to express a feature request. Currently, the shex files that act as input need to be put into a designated directory. However, Wikidata also has the EntitySchema extensions where schemas can be identified by the Eid.
Would it be possible to feed wdsub an eid where the script would fetch the latest schema from the EntitySchema extension? Did would allow a more straightforward running of the script since less time is spent on authoring schema to describe the boundaries of the subset

[labra]

Yes, that sounds a nice feature. It depends on issue #19