Structure the references (normative, informative)

[Ostrzyciel]

I've recently been pointed to this: https://openengiadina.codeberg.page/rdf-cbor/

It looks very interesting... with some of the same goals as Jelly, but also some different ones. I think what is most relevant for us here is that there is something to learn from this spec.

Although Jelly is based on Protobuf which is not (and I doubt ever will be) a standard, it would make sense to have a references section, split into normative and informative references: https://openengiadina.codeberg.page/rdf-cbor/#name-references

Then, it would be clearer what documents are we referring to in the spec. We'd need to look into how to render such references with mkdocs.

[Ostrzyciel]

We could use the formatting approach from this spec: https://rdf-tensor.github.io/jena-datatensor/