[rdf] Serialize queries as Hydra collections

[ddeboer]

• Add a controller at /hydra/123 that serializes Zotonic queries as Hydra collections, either in JSON-LD or Turtle.
• Hydra gives machine-readable pagination over large collections. See https://www.hydra-cg.com/spec/latest/core/.
• Filter out empty triple objects in both JSON-LD and Turtle serializations. This makes it easy to dump some triples that may not have a value, such as previous or next page in our case.
• The /hydra endpoint can be used as the RDF distribution of datasets, which are a subcategory of query.
• Add documentation.