fix(office-ui): stop progress-row flicker and surface native role replies end to end

Two user-visible defects in company mode, root-caused via project 6666/8888
DB forensics:

Progress-row flicker (thinking preview appearing/disappearing): progress
entries were broadcast to clients before reaching the persistence buffer,
so a tool_call-triggered session_detail snapshot rebuilt from the DB erased
freshly streamed entries from the live log. Buffer now fills before the
broadcast and session_detail flushes it before reading.

Native role transcripts incomplete (thinking only at start, no narration,
no final summary — external agents unaffected):
- thinking deltas shared one stream id per conversation turn while seq
  reset per iteration, collapsing all iterations into one entry and
  silently dropping live thinking from iteration 2 on; now keyed per
  iteration like assistant deltas
- assistant_delta events were mapped to None; company mode now surfaces
  them as streaming 'assistant' progress entries (rendered as Reply cards,
  merged like thinking, excluded from inline chat rows)
- thinking was persisted one row per token, flooding the 1000-entry cap
  and evicting interleaved tool history; append_progress now folds
  streaming deltas per (type, turn, stream) with seq dedup
- the terminal company turn was hidden at summary detail and, worse, its
  id-keyed backfill merge kept the first-inserted intermediate content, so
  the final reply never reached any channel; terminal turns are now
  flagged company_final_turn, visible at summary detail, and carry their
  own ui_message_id so they insert as fresh rows
- appendProgressEntry applied its seq guard against unrelated entries when
  the stream key was absent from the log, killing the first delta of any
  fresh stream; the guard now only applies within the same stream

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

opc/layer3_agent/runtime_v2/runtime.py

@@ -435,8 +435,13 @@ class NativeRuntimeV2:
                                     "canonical_turn_id": conversation_turn_id,
                                     "conversation_turn_id": conversation_turn_id,
                                     "execution_turn_id": execution_turn_id,
-                                    "item_id": f"{turn_id}:thinking",
-                                    "stream_id": f"{turn_id}:thinking",
+                                    # Keyed per iteration (like assistant_delta):
+                                    # thinking_delta_seq resets every iteration, so a
+                                    # turn-scoped stream id makes downstream seq guards
+                                    # drop iteration>=2 deltas and collapses all
+                                    # iterations into one entry (no tool interleaving).
+                                    "item_id": f"{execution_turn_id}:thinking",
+                                    "stream_id": f"{execution_turn_id}:thinking",
                                     "seq": thinking_delta_seq,
                                     "text": thinking_text,
                                 },
@@ -2797,6 +2802,11 @@ class NativeRuntimeV2:
         if is_company_mode:
             metadata["execution_mode"] = "company_mode"
             metadata["company_runtime_raw_turn"] = True
+            if not tool_calls:
+                # Terminal iteration of the company turn — this is the role's
+                # final reply. Marked so the UI can show it at summary detail
+                # even though the kind is otherwise full-detail-only.
+                metadata["company_final_turn"] = True
             if task.assigned_to:
                 metadata["role_id"] = str(task.assigned_to)
         else:
@@ -2808,7 +2818,15 @@ class NativeRuntimeV2:
             if message_turn_id and message_turn_id != canonical_turn_id:
                 metadata["execution_turn_id"] = message_turn_id
             if is_company_mode:
-                metadata["ui_message_id"] = f"runtime-v2-company-assistant:{canonical_turn_id}"
+                # Tool-calling iterations of a company conversation turn share
+                # one UI row; the terminal reply gets its own id. The id-keyed
+                # backfill merge keeps the first-inserted content for same-kind
+                # candidates, so reusing the shared id freezes the row at
+                # iteration 1 and swallows the final reply entirely.
+                if metadata.get("company_final_turn"):
+                    metadata["ui_message_id"] = f"runtime-v2-company-assistant-final:{canonical_turn_id}"
+                else:
+                    metadata["ui_message_id"] = f"runtime-v2-company-assistant:{canonical_turn_id}"
             elif is_intermediate_tool_turn:
                 metadata["ui_message_id"] = f"runtime-v2-intermediate-assistant:{message_turn_id or canonical_turn_id}"
             else: