Calculate `log_prior` when constructing result posterior

The current result posterior output by parallel-bilby (using Dynesty) is lacking of the log_prior evaluation (result.posterior[‘log_prior’] returns an array of None).

This is because the result.samples_to_posterior function requires an explicit input of priors to trigger the calculations of log_prior, (e.g.: bilby.core.sampler.__init__.py)

This PR attempts to solve this issue by adding priors=result.priors to the arguments when calling the function.