Improve the models to work with the current api version. add tests
that use fake-data, that was captured from a real vCloud.
Also removed a couple of dead code. Tests have been added for
everything that changed or have been added. All the existing but
untouched parts have not been tested.
After reading the API doc, in my opinion networks work different than
currently implemented here. To get the basic stuff working, we remove
all the code with unkwown implementation.
Changing a few bits and bytes here and there. This breaks some apis,
but allows us to get more information about the VM and addresses
also the various parts that didn't work so far.
A vcloud usually contains more than one organization which then
contains all the other computing parts (such as vdcs etc.). So far
we keep the old way, to query the first organization by default, but
also allow it to be set.
Many changes:
-Become more of a proper Fog Service.
-Drop most of the Collection/Model customizations.
-Add a hook for services to do something after #new with the instance
-Move to the ToHashDocument parser ... so no more having to actually
make parsers.
-Fog::Vcloud::Extension is kind of like Fog::Service, but for writing
extension modules to Fog::Vcloud.
-Fix up existing specs/mocks (they're not complete atm, but the existing
ones are up to date).
-Fog::Vcloud::Terremark::Ecloud gets almost all extensions implemented
(almost).
-Fog::Vcloud::Terremark::Ecloud bumped to working with the current TMRK
API release.
-Factor out some TMRK/ecloud specifc mock data into the ecloud module.
-Probably forgetting something.
