Simulating recombination

[ArtPoon]

This requires two new functions that are applied to an InnerTree:

1. Sample recombination events that occur at a constant rate over time, conditional on the presence of two or more lineages that co-exist within the same Host. (The rate is therefore multiplied by the number of co-existing pairs.) This results in an ancestral recombination graph (ARG).
2. Assign breakpoint locations for recombination events and resolve the ARG into a sequence of trees.

[ArtPoon]

I think the above might be wrong. A recombination event should split a lineage, so it can affect a single lineage within a host (we don't need to have two or more). Note that we are tracking lineages that are ancestral to a sample.

We are also going to assume that recombination and coalescence cannot occur simultaneously and that they are independent events. That makes recombination a competing process that occurs at some rate along with the other kinds of events (e.g., coalescence).