Add use cases based on current NDJSON-LD usage

[Ostrzyciel]

The JSON-LD WG has discussed in the past something called NDJSON-LD that allows for having multiple JSON-LD document in a single file, by separating them with newlines: https://github.com/json-ld/ndjson-ld

This work was not continued, and the current JSON-LD WG charter (until 31 January 2028) does not foresee working on it.

However, it was used in practice. From the JSON-LD WG call I got the following pointers:

• We (National Library of Sweden) use ndjson as a dump format (full dumps then kept in sync using https://emm-spec.org/1.0/ )
• <frenchrh_> We use JSON-LD for FAIRifying Scientific Data for acquistion, analysis, modeling
• <frenchrh_> And we have a Materials DAta Science Ontology (MDS-Onto, https://cwrusdle.bitbucket.io/ )
• See also ND-delimited JSON-LD in RDF4J: Introduce RDFParser and RDFWriter implementation for Newline Delimited JSON-LD format eclipse-rdf4j/rdf4j#2840

I think this is valuable. We should compile this into a use case mention in the spec.

[jpcik]

Maybe have a discussion with people involved?

[Ostrzyciel]

In #14 I added the first draft of the NDJSON-LD format to the spec.