Currently we run the serialization 2 times to first generate new work items and then regenerate all of them in a second round to make sure that there are no references to newly created work items missing
We should on the first run create stub work items with the capella UUID and the title only and run the real full serialization after posting missing work items
Currently we run the serialization 2 times to first generate new work items and then regenerate all of them in a second round to make sure that there are no references to newly created work items missing We should on the first run create stub work items with the capella UUID and the title only and run the real full serialization after posting missing work items