Default prior bounds

See merge request Monash/tupak!34

Allow multiple prior draws

See merge request Monash/tupak!33

- Rename n_samples to size to fit with standard numpy conventions
- Fix check to handle multiple values

Functionally, this changes nothing, but it makes it clear that the
filled priors are the dictionary `priors`.

Previously, the dynesty kwarg for the number of live points was set to
be `npoints`, and so `npoints=10000` was passed to
`dynesty.NestedSampler`. This, however, is not the kwarg requried by
`NestedSampler`, but it instead should be `nlive`.  Unfortunately, this
error is silenced because the kwargs are just lost in dynesty.

- Fixes nlive kwarg
- Reduces default number from 10000 to 250 (as this was what was
  actually used before)
- Adds a verbose flag controlling the `run_nested` argument
  `print_progress`

Add minimal example and associated helper functions

See merge request Monash/tupak!32

- Adds an example showing the simplest number of steps to run PE on
  GW150914
- Adds a helper (rapper) function to quickly get the BBH likelihood
  instance

- Add ability to pass in parameters as either keys or latex labels
- Update truth to not overwrite the given kwarg, but preserve "plot
  injections by default" behaviour.
- Add function to convert a set of parameter_keys into the latex labels
  used by CC

- Set plot=False during tests
- Remove verbose=True from nestle calls

1) Remove the prior from the result object: we already save them as a
text file, so I don't think they need to be in the result. For now, I've
only commented them out with a message as there may be someone tempted
in the future to add it back in.

2) If the saving data does error, it now continues (without writing the
data) and prints a big error message.