Confusing kwarg in Nested

[ParadaCarleton]

Is this the maximum or minimum?

[mileslucas]

Ah, it should read minimum. It could use a better description, too. Something like,

"Minimum efficiency (samples accepted / samples generated) before fitting the first bounding volume. This is to allow some burn-in of the sampler and avoid spuriously shrinking the bounds, which can decrease evidence estimate accuracy."

[ParadaCarleton]

Ah, it should read minimum. It could use a better description, too. Something like,

"Minimum efficiency (samples accepted / samples generated) before fitting the first bounding volume. This is to allow some burn-in of the sampler and avoid spuriously shrinking the bounds, which can decrease evidence estimate accuracy."

Would you endorse these docstrings?

- `min_ncall`: The minimum number of iterations before fitting the first bound; used to 
avoid shrinking the bounds before burn-in is completed.
- `min_eff`: Minimum efficiency `(samples accepted / samples generated)` before fitting the 
first bound; used to avoid shrinking the bounds before burn-in is completed.

[mileslucas]

Yes those look great- feel free to PR!