coalescent_rate_trajectory and friends assumes diploidy

[petrelharp]

As pointed out in #1761, these functions assume diploidy. Fixing this would just require adding a ploidy argument that replaces the 2 right here.

[jeromekelleher]

My take is that this is an enhancement, since ploidy=2 is the default elsewhere, so adding a ploidy argument is a strictly compatible change.

If someone would like to take this on, I think the main work would be validating the output for different ploidy levels, making sure to test for things population size changes as shown in #1761.