Made Agent+DM a bit more robust, where if finding the best goal's goal plan or goal plan variant doesn't pull through and fail, it will go on and try to find the next best goal.