Make checkpoint interval as expected

People may have noticed that the checkpoint interval continues throughout a run. I'm not sure if this is down to a misunderstanding of mine or a change in dynesty, but it's been an issue for a long time.

With this change, the checkpoint will actually be written every n_check_point, rather than the current (n(n+1)/2 - 1) * n_check_point.

If people like the current behaviour, we should update the documentation.

added Sampling label

approved this merge request

approved this merge request

mentioned in commit parallel_bilby@e0036954

mentioned in merge request parallel_bilby!47 (merged)

changed milestone to %0.6.6

mentioned in commit 2c7dd519

merged

mentioned in commit parallel_bilby@eb968e50