docs(vtags): note rename-respawn transient + write-side-only historize (integration review)

src/Server/ZB.MOM.WW.OtOpcUa.OpcUaServer/Phase7Applier.cs

@@ -253,6 +253,10 @@ public sealed class Phase7Applier
 
         // Variables: NodeId is FOLDER-SCOPED ("<parent>/<Name>"), mirroring the equipment-tag pass.
         // Parent is the FolderPath sub-folder when set, else the equipment folder directly.
+        // NOTE (H5): a VirtualTag's Historize flag is honoured on the WRITE side only — VirtualTagHostActor
+        // forwards historized results to IHistoryWriter. It is intentionally NOT materialised as an SDK
+        // Historizing/HistoryRead variable here (no server-side OPC UA HistoryRead for vtags), so these
+        // stay plain read-only computed-output nodes.
         foreach (var v in composition.EquipmentVirtualTags)
         {
             var parent = string.IsNullOrWhiteSpace(v.FolderPath)

src/Server/ZB.MOM.WW.OtOpcUa.Runtime/VirtualTags/VirtualTagHostActor.cs

@@ -113,6 +113,11 @@ public sealed class VirtualTagHostActor : ReceiveActor
         // Rebuild the NodeId map every apply so renames (Name/FolderPath/EquipmentId changes) are
         // picked up. The map only contains currently-desired vtags, so a result for a removed vtag
         // finds no entry and is dropped.
+        // NOTE: if a respawn (above) was triggered by a Name/FolderPath/EquipmentId change, the
+        // stopping old child may still deliver ONE in-flight result keyed by the same vtagId, which
+        // OnResult would publish to the NEW NodeId. This is a transient phantom that the new child's
+        // next evaluation overwrites — acceptable (and strictly better than the pre-H1b behaviour,
+        // where a renamed vtag was a silent no-op).
         _nodeIdByVtag.Clear();
         foreach (var p in msg.Plans)
         {