refactor(journal): fix flush, token tracking, and consolidate tests

RunJournal fixes:
- _flush_sync: retain events in buffer when no event loop instead of
  dropping them; worker's finally block flushes via async flush().
- on_llm_end: add tool_calls filter and caller=="lead_agent" guard for
  ai_message events; mark message IDs for dedup with record_llm_usage.
- worker.py: persist completion data (tokens, message count) to RunStore
  in finally block.

Model factory:
- Auto-inject stream_usage=True for BaseChatOpenAI subclasses with
  custom api_base, so usage_metadata is populated in streaming responses.

Test consolidation:
- Delete test_phase2b_integration.py (redundant with existing tests).
- Move DB-backed lifecycle test into test_run_journal.py.
- Add tests for stream_usage injection in test_model_factory.py.
- Clean up executor/task_tool dead journal references.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

backend/tests/test_model_factory.py

@@ -593,6 +593,84 @@ def test_codex_provider_strips_unsupported_max_tokens(monkeypatch):
     assert "max_tokens" not in FakeChatModel.captured_kwargs
 
 
+# ---------------------------------------------------------------------------
+# stream_usage injection
+# ---------------------------------------------------------------------------
+
+
+class _FakeWithStreamUsage(FakeChatModel):
+    """Fake model that declares stream_usage in model_fields (like BaseChatOpenAI)."""
+
+    stream_usage: bool | None = None
+
+
+def test_stream_usage_injected_for_openai_compatible_model(monkeypatch):
+    """Factory should set stream_usage=True for models with stream_usage field."""
+    cfg = _make_app_config([_make_model("deepseek", use="langchain_deepseek:ChatDeepSeek")])
+    _patch_factory(monkeypatch, cfg, model_class=_FakeWithStreamUsage)
+
+    captured: dict = {}
+
+    class CapturingModel(_FakeWithStreamUsage):
+        def __init__(self, **kwargs):
+            captured.update(kwargs)
+            BaseChatModel.__init__(self, **kwargs)
+
+    monkeypatch.setattr(factory_module, "resolve_class", lambda path, base: CapturingModel)
+
+    factory_module.create_chat_model(name="deepseek")
+
+    assert captured.get("stream_usage") is True
+
+
+def test_stream_usage_not_injected_for_non_openai_model(monkeypatch):
+    """Factory should NOT inject stream_usage for models without the field."""
+    cfg = _make_app_config([_make_model("claude", use="langchain_anthropic:ChatAnthropic")])
+    _patch_factory(monkeypatch, cfg)
+
+    captured: dict = {}
+
+    class CapturingModel(FakeChatModel):
+        def __init__(self, **kwargs):
+            captured.update(kwargs)
+            BaseChatModel.__init__(self, **kwargs)
+
+    monkeypatch.setattr(factory_module, "resolve_class", lambda path, base: CapturingModel)
+
+    factory_module.create_chat_model(name="claude")
+
+    assert "stream_usage" not in captured
+
+
+def test_stream_usage_not_overridden_when_explicitly_set_in_config(monkeypatch):
+    """If config dumps stream_usage=False, factory should respect it."""
+    cfg = _make_app_config([_make_model("deepseek", use="langchain_deepseek:ChatDeepSeek")])
+    _patch_factory(monkeypatch, cfg, model_class=_FakeWithStreamUsage)
+
+    captured: dict = {}
+
+    class CapturingModel(_FakeWithStreamUsage):
+        def __init__(self, **kwargs):
+            captured.update(kwargs)
+            BaseChatModel.__init__(self, **kwargs)
+
+    monkeypatch.setattr(factory_module, "resolve_class", lambda path, base: CapturingModel)
+
+    # Simulate config having stream_usage explicitly set by patching model_dump
+    original_get_model_config = cfg.get_model_config
+
+    def patched_get_model_config(name):
+        mc = original_get_model_config(name)
+        mc.stream_usage = False  # type: ignore[attr-defined]
+        return mc
+
+    monkeypatch.setattr(cfg, "get_model_config", patched_get_model_config)
+
+    factory_module.create_chat_model(name="deepseek")
+
+    assert captured.get("stream_usage") is False
+
+
 def test_openai_responses_api_settings_are_passed_to_chatopenai(monkeypatch):
     model = ModelConfig(
         name="gpt-5-responses",