- Adds a `result_class` argument to run_sampler which can determine
  which result class to use
- Extend the `meta_data` use in the gw module to save more information
- Adds a bilby.gw.result.CBCResult class which unpacks the meta_data to
  provide easy access to commonly required information

Add another exception to catch in `plot_with_data`

See merge request lscsoft/bilby!329

Resolve "Enable different query types to be passed to gw_data_find command line"

Closes #268

See merge request lscsoft/bilby!328

Adds a simple print out of the label and outdir before commencing

See merge request lscsoft/bilby!326

For scripts with multiple `run_sampler` commands, it isn't always clear
which stage one is at. This clearly prints that out and will be useful
for debugging.

Resolve "`ModuleNotFoundError` exception doesn't exist in python2"

Closes #263

See merge request lscsoft/bilby!318

Include license in tarballs and wheels

See merge request lscsoft/bilby!323

Fix pymultinest failing when using long name

See merge request lscsoft/bilby!321

Build from source ubuntu distribution.

adding additional information when performing redundancy check, to clarify…

See merge request lscsoft/bilby!309

Add loglikelihood evaluations to pymultinest

See merge request lscsoft/bilby!310

Comment in GW150914.py

See merge request lscsoft/bilby!314

Adds close calls after savefig

See merge request lscsoft/bilby!316

Adds a simple hack to ensure corner doesn't fail

See merge request lscsoft/bilby!313

Update CONTRIBUTING.md with specific fork/merge instructions

See merge request lscsoft/bilby!290

If there are NaNs in the posterior, continue onward but print a message

See merge request lscsoft/bilby!312

Adding a comment in GW150914.py for running pe on other events available on LOSC, i.e. for GW170817 the kwarg "tag" is required to download the data.

Resolve "Discrepancy in definitions of matched filter SNR"

Closes #255

See merge request lscsoft/bilby!306

update with catalog events

See merge request lscsoft/bilby!307

NaNs in the posterior are generted by the conversion function (say when
the prior is fixed to zero for the spins). This causes the
plot_marginals to fail. Instead, a message will be printed, but the
script will otherwise continue.

If the prior is fixed at a constant, but you include the parameter name
in the parameters given to plot_corner. It currently fails with a
confusing message about the range. This just sets the range which means
a plot will be created.

Once the run is complete, read in the log likelihood from the
post_equal_weights.dat file and store it. This also reads in the samples
from the same place to avoid any mismatch in ordering between the two.
For a description of the post_equal_weights file see
https://github.com/JohannesBuchner/MultiNest

add conversions documentation

See merge request lscsoft/bilby!308

adding additional information when performing redundancy check, to clarify confusing BBHPriorDict output when using BNSPriorDict

fix ordering after interpolation

See merge request lscsoft/bilby!301

Resolve "Prior read in from file chokes on empty lines"

Closes #265

See merge request lscsoft/bilby!303

Resolve "copying BBHPriorDict"

Closes #258

See merge request lscsoft/bilby!305