Distinguish "not coded" from "not codable" - once COS data is integrated

[LukasWallrich]

With COS data, we will have entries where success criteria are confirmed as not (feasibly) codable - should then distinguish that from not coded? (Maybe - shows our effort better, but makes a claim that can be challenged ... tbd)