Restructre time frequency tests and fixing some recently surfaced bugs when using time domain models

See merge request !279

Adds two new methods to the result object: kde and posterior_probability

See merge request !278

Resolve "Make Interferometers serialisable and add a to_hdf5 method"

Closes #242

See merge request !273

Resolve "PEP8 Imports"

Closes #207

See merge request !230

Resolve #222 Inconsistent use of `priors` and `prior`

Closes #222

See merge request !274

Adds version information to the result object

See merge request lscsoft/bilby!275

1) Rename 'path' to outdir
2) Add detector names to the interferometer list filename
3) Make from_hdf5 method only be given a filename for simplicity

Merge remote-tracking branch 'origin/master' into 242-make-interferometers-serialisable-and-add-a-to_hdf5-method

These can be used to estimate a posterior probability at some new point
given the set of samples stored in the posterior.

Bug fix time marginalisation

See merge request lscsoft/bilby!276

Merge remote-tracking branch 'origin/master' into 242-make-interferometers-serialisable-and-add-a-to_hdf5-method

# Conflicts:
#	bilby/gw/detector.py

Encapsulate the lalsimulation function calls

See merge request !266

Resolve "Create single class that handles `duration`, `sampling_frequency`, `start_time`, `frequency_array`, `time_array`, and maybe `frequency_domain_strain`, `time_domain_strain`"

Closes #227

See merge request !257

Resolve "Create single class that handles `duration`, `sampling_frequency`, `start_time`, `frequency_array`, `time_array`, and maybe `frequency_domain_strain`, `time_domain_strain`"