Label sequence groups and tree internal nodes with diversity information

Work on JAL-4386 demonstrated creating multiple nodes for a sequence, one for each secondary structure annotation track, is useful when exploring the differences between different sources of truth for secondary structure (ie experiment, AlphaFold, SWISS-MODEL, etc) across a family of proteins.

Jalview's tree viewer and the sequence groups it creates could be enhanced to report summaries of those sources of truth - e.g. one branch only contains alphafold predictions, another only contains PDB and SWISS-MODEL predictions.