When ready, add a section on "branch length" equivalents of e.g. SFS measures

[hyanwong]

Just noting this here - I think Peter Ralph has code to do this incrementally and efficiently over a tree sequence. Once it is imported into the tskit base, we should add an example (e.g. using the SLiM dataset) to show a plot of selection having an effect on the trees.