v6.9.7: Release notes: correct 6.9.7 notes, drop stale FMMM line from 6.9.6

SplitsTree 6.9.7

New features

• network view: a Straighten button redraws a network on a grid, made rectilinear where possible
• network view: a grid toggle snaps the nodes to a grid, so a drawing can be tidied up by hand
• Group Identical Haplotypes now names the new document after the source, e.g. foo -> foo-haplotypes
• Simplify Filter: the automatic target is now called Balanced and is the default, so a newly added filter
  simplifies straight away; the damage measure is renamed Deviation (workflows saved by 6.9.6 still load)
• the alignment viewer now recognises DNA/RNA data that carries ambiguity codes as nucleotide data, so it defaults
  to the Nucleotide colouring and allows codon selection
• algorithm tabs can show a separator between groups of options
• As Network..., in the context menu of a split network view or a single-tree tree view, opens the drawing as
  an editable network in a new document (foo -> foo-network), keeping the coordinates it is currently drawn with

Bug fixes

• in the alignment viewer, the list of taxa (left) and the sequences (right) could become detached and stop
  scrolling together; they now stay in sync
• a DNA/RNA alignment whose only symbol beyond A, C, G, T (or U) is N is now read as DNA (RNA) with N as the
  missing character, instead of as a DNAwithAmbiguityCodes type that listed N among its symbols
• network view: dragging an unselected node no longer drags the selected nodes; a drag moves the selection only
  when it starts on a selected node
• network view: each vertex label is written only once, so saving and reloading a network no longer repeats it

Known issues

• LogDet requires every state to be observed within a pair of sequences, so it stays undefined for most pairs
  of a strongly skewed protein alignment