Add support for generalized RDF input

[adanilenka]

Currently, input files with generalized RDF statements are not supported even when --opt.generalized-statements=true

For example file, refer to this test case: https://github.com/Jelly-RDF/jelly-protobuf/tree/main/test/rdf/to_jelly/triples_rdf_1_1/neg_001

Or use the following set of triples:

<http://example.org/resource/r1> _:b1 <http://example.org/resource/r2> .
"Resource 1" <http://example.org/property/p> <http://example.org/resource/r3> .
<http://example.org/resource/r3> "Property Label" <http://example.org/resource/r1> .

Executing command:
jelly-cli rdf to-jelly ./neg_001/in_001.nt --to ./neg_001/out.jelly --opt.generalized-statements=**true** --opt.rdf-star=false --opt.max-name-table-size=8 --opt.max-prefix-table-size=0 --opt.max-datatype-table-size=4

Leads to the following error:

This feature is required for test cases that include generalized RDF.

[Ostrzyciel]

That's a tough one, because Jena doesn't really support parsing generalized N-Triples files: apache/jena#3079

It does support parsing binary generalized RDF, though. It can also write generalized N-Triples.

I'll see what I can do here.

[Ostrzyciel]

@adanilenka please paste NT files and other stuff like this in code blocks (this thingy to start and end: ```)

Otherwise the <> characters get lost and I can't copy the code easily.