Duplicate detection should not erase document history

Duplicate detection is usually framed as an efficiency feature. That makes sense, but it can create a subtle trust problem if it suppresses repeated documents without preserving the case history aroun