Making glob() available in session/configuration.py allows users of
savefluxes and sfhfromfile to use something like

filename = "eval glob('sfh*.fits')"
sfr_column = 2

to take the star formation history from several files.

Remove the author email addresses from the source files and add a AUTHORS.rst
file with the addresses and current affiliation to make it easier to update
the information.

Also, give copyright to the University of Cambridge and the University of
Crete for code developed by authors working there.

For the creation and analysis modules, move the contents of the
common.py files to the __init__.py.

Move the SED creation modules to pcigale.creation_modules to be more
consistent with the analysis module location.

For redshift and analysis module, we don't need to evaluate each
parameter as an array as we don't expect arrays here (and that force us
to use parameter[0] to get the value).

Configobj return strings, to evaluate the booleans on their "string"
content.

In order to speed-up module re-use, get the module needs (ssp,
filters...) out of pcigale database at module instantiation instead of
doing this everytime "process" is called.

To do that:
- we must give the module parameters (and proceed with their checking)
at
  module instantiation.
- we must add the possibility for each module to have its own
  initialisation code.

In addition, to create the configuration file, we must be able to load
the modules without any parameter to query them for their parameter
list.

The bc03, dh2002 and m2005 modules were changed to use this new
behavior.

The redshifting module must be separated from the main modules producing
the restframe SED, because the redshift parameter is taken directly from
the observation list.
Nevertheless, it is still a SED creation module because one may want to
use it for testing various redshifts.

s/[Pp]arametre/[Pp]arameter/g
s/[Dd]ictionnary/[Dd]ictionary/g

When a numpy array was evaluated, it was interpreted as a list of
possible values (although we generaly want the parametre value to be the
array itself).

Also, we changed the way to identify to-be-evaluated strings. The now
must begin with "eval " and are expected to return an array.

read_array_description was renamed to evaluate_description and modified
to take info account the fact that configobj returns arrays and that the
parametre list may not be a numpy array.