Error management #25
Description
We need to improve how errors are handled by wimpy. At the moment, if stops executing whenever a task fails. Since many of the things wimpy creates are part of a CloudFormation stack, it's not a big problem if it fails during the CloudFormation creation, but if it fails before/after that, we could stop in some inconsisten state. For example, we could create a Github Deployment Status object for the pending state, but never actually changing the status to success/failure.