Update Hub IDs to the Qdrant org

docs: remove repository demo link

docs: link runnable Constella demo

docs: keep benchmark reporting focused on established results

docs: note full BEIR benchmark in progress

docs: explain the Constella family and recommended retrieval setup

docs: simplify and align model card

Update model card for the research preview: one FastEmbed branch for the whole family, native fp32 query vectors, canonical clean-4 results with all six beside them.

Upload constella-nano research preview

initial commit