PE paper:
https://dcc.ligo.org/public/0122/P1500218/014/PhysRevLett.116.241102.pdf

# Conflicts:
#	peyote/detector.py

When setting data, gives the user the option to provide the epoch (the
GPS time at which the data begins). This is then used in the time-shift.

- Rewrites the phase shift operation
- Adds epoch as option when setting the data
- Renames epoch -> center_time in get_interferometer to avoid confusion
- Fix up the documentation of set_data

pymultinest output is now placed in a subdirectory of outdir demarked by
the label. This ensures it won't be in conflict with other scripts using
the same outdir.

- Demonstrated in GW150914.py and BasicTutorial.py

Adds completed version of the det_interfometer function which. Create an
IFO instance with a PSD and data appropriate for data analysis. Example
usage is given in the GW150914.py script.

Note - a broken version was accidently added in the previous commit.

Puts the creation of inteferometers into a function. In theory, this
means we don't need to instantiate them all at `import`, but for
backward compatibility I have included them.

Primary issues that this resolves:
- Include many different variables in the PE rather than a small subset
- Use only 100s to estimate the PSD
- Use an alpha=0.25 Tukey window (matched to data generated by
  lalinference)
Closes #23

Moritz Huebner: Fixed with new prior class and removed ra, dec, ... parameters from the source function definitions.

Unify parameter and prior classes

See merge request Monash/peyote!17