Add Occam factor

Closes #93

See merge request Monash/tupak!86

Fixed a bug in create_time_series that produced incorrect time_arrays if a starting time other than 0 is given

Fixed a bug in create_time_series that produced incorrect time_arrays if a starting time other than 0 is given

Fixed a bug in create_time_series that produced incorrect time_arrays if a starting time other than 0 is given

Moves the plot_walker method to the Results object. The benefit here is
that we can generate walker plots after the run and mitigate the chances
of a plot error spoiling a run.

- Also adds ability pass through the 'a' parameter used by emcee

Make the corner plots use 1, 2, and 3-sigma countours

See merge request Monash/tupak!89

Based on http://corner.readthedocs.io/en/latest/pages/sigmas.html

Adds the emcee sampler class

See merge request Monash/tupak!88

Regularize default estimation of n_check_point

Closes #117

See merge request Monash/tupak!87

Make n_check_point estimation round to the nearest order of magnitude.

Fixes #117

This adds methods to __dir__ so that plot_corner and other methods can
be tab completed (in ipython). Unfortunately, this is done by setting
these directly. While this isn't an ideal solution - there is no obvious
way to automate the process as methods to looks up all methods
themselves use the __dir__ attribute causing a recursion error.

Closes #111

fix conversion tests

See merge request Monash/tupak!85

Tidy up of inject signal

Closes #115 and #116

See merge request Monash/tupak!82

Sets the filter_frequency used in frame_file based on the maximum

Closes #108

See merge request Monash/tupak!83

- Provides backward compatibility for people with scripts using the old
  style
- Also because, as Colm pointed out, there may be bugger behaviour with
  multiple detectors and it may take a long time to calculate the waveform
- Change the name in `inject_signal` from
  waveform_polarizations->injection_polarizations.

This last step could break peoples code who are using inject_signal
directly, but its probably best to unify notation now rather than have
two names for one thing.

This reverts commit 2ab4bddd.

A non-ideal fix to get the coverage badge to work. Seemingly it was
failing before because it now sees tupak as being in the dist-packages.
I've added a hard path to that directory so that at least the numbers
are correct.