feat(onboarding): contextual first-touch hints for /busy and /verbose (#16046)

Instead of a blocking first-run questionnaire, show a one-time hint the first
time the user hits each behavior fork:

1. First message while the agent is working — appends a hint to the busy-ack
   explaining the /busy queue vs /busy interrupt knob, phrased to match the
   mode that was just applied (don't tell a queue-mode user to switch to
   queue).

2. First tool that runs for >= 30s in the noisiest progress mode
   (tool_progress: all) — prints a hint about /verbose to cycle display
   modes (all -> new -> off -> verbose). Gated on /verbose actually being
   usable on the surface: always shown on CLI; on gateway only shown when
   display.tool_progress_command is enabled.

Each hint is latched in config.yaml under onboarding.seen.<flag>, so it
fires exactly once per install across CLI, gateway, and cron, then never
again. Users can wipe the section to re-see hints.

New:
- agent/onboarding.py — is_seen / mark_seen / hint strings, shared by
  both CLI and gateway.
- onboarding.seen in DEFAULT_CONFIG (hermes_cli/config.py) and in
  load_cli_config defaults (cli.py). No _config_version bump — deep
  merge handles new keys.

Wired:
- gateway/run.py: _handle_active_session_busy_message appends the hint
  after building the ack.  progress_callback tracks tool.completed
  duration and queues the tool-progress hint into the progress bubble.
- cli.py: CLI input loop appends the busy-input hint on the first busy
  Enter; _on_tool_progress appends the tool-progress hint on the first
  >=30s tool completion.  In-memory CLI_CONFIG is also updated so
  subsequent fires in the same process are suppressed immediately.

All writes go through atomic_yaml_write and are wrapped in try/except
so onboarding can never break the input/busy-ack paths.

tests/gateway/test_busy_session_ack.py

@@ -349,3 +349,121 @@ class TestBusySessionAck:
 
         result = await runner._handle_active_session_busy_message(event, sk)
         assert result is False  # not handled, let default path try
+
+
+class TestBusySessionOnboardingHint:
+    """First-touch hint appended to the busy-ack the first time it fires."""
+
+    @pytest.mark.asyncio
+    async def test_first_busy_ack_appends_interrupt_hint(self, tmp_path, monkeypatch):
+        """First busy-while-running message gets an extra hint about /busy."""
+        import gateway.run as _gr
+
+        monkeypatch.setattr(_gr, "_hermes_home", tmp_path)
+        # mark_seen imports utils.atomic_yaml_write; make sure it resolves
+        # against a writable dir by pointing _hermes_home at tmp_path.
+        monkeypatch.setattr(_gr, "_load_gateway_config", lambda: {})
+
+        runner, _sentinel = _make_runner()
+        runner._busy_input_mode = "interrupt"
+        adapter = _make_adapter()
+
+        event = _make_event(text="ping")
+        sk = build_session_key(event.source)
+
+        agent = MagicMock()
+        agent.get_activity_summary.return_value = {
+            "api_call_count": 3, "max_iterations": 60,
+            "current_tool": None, "last_activity_ts": time.time(),
+            "last_activity_desc": "api", "seconds_since_activity": 0.1,
+        }
+        runner._running_agents[sk] = agent
+        runner._running_agents_ts[sk] = time.time() - 5
+        runner.adapters[event.source.platform] = adapter
+
+        await runner._handle_active_session_busy_message(event, sk)
+
+        call_kwargs = adapter._send_with_retry.call_args
+        content = call_kwargs.kwargs.get("content", "")
+
+        # Normal ack body
+        assert "Interrupting" in content
+        # First-touch hint appended
+        assert "First-time tip" in content
+        assert "/busy queue" in content
+
+        # The flag is now persisted to tmp_path/config.yaml
+        import yaml
+        cfg = yaml.safe_load((tmp_path / "config.yaml").read_text())
+        assert cfg["onboarding"]["seen"]["busy_input_prompt"] is True
+
+    @pytest.mark.asyncio
+    async def test_second_busy_ack_omits_hint(self, tmp_path, monkeypatch):
+        """Once the flag is marked, the hint never appears again."""
+        import gateway.run as _gr
+        import yaml
+
+        monkeypatch.setattr(_gr, "_hermes_home", tmp_path)
+        # Pre-populate the config so is_seen() returns True from the start.
+        (tmp_path / "config.yaml").write_text(yaml.safe_dump({
+            "onboarding": {"seen": {"busy_input_prompt": True}},
+        }))
+        monkeypatch.setattr(
+            _gr, "_load_gateway_config",
+            lambda: yaml.safe_load((tmp_path / "config.yaml").read_text()),
+        )
+
+        runner, _sentinel = _make_runner()
+        runner._busy_input_mode = "interrupt"
+        adapter = _make_adapter()
+
+        event = _make_event(text="ping again")
+        sk = build_session_key(event.source)
+
+        agent = MagicMock()
+        agent.get_activity_summary.return_value = {
+            "api_call_count": 3, "max_iterations": 60,
+            "current_tool": None, "last_activity_ts": time.time(),
+            "last_activity_desc": "api", "seconds_since_activity": 0.1,
+        }
+        runner._running_agents[sk] = agent
+        runner._running_agents_ts[sk] = time.time() - 5
+        runner.adapters[event.source.platform] = adapter
+
+        await runner._handle_active_session_busy_message(event, sk)
+
+        call_kwargs = adapter._send_with_retry.call_args
+        content = call_kwargs.kwargs.get("content", "")
+
+        assert "Interrupting" in content
+        assert "First-time tip" not in content
+        assert "/busy queue" not in content
+
+    @pytest.mark.asyncio
+    async def test_queue_mode_hint_points_to_interrupt(self, tmp_path, monkeypatch):
+        """In queue mode the hint should suggest /busy interrupt, not /busy queue."""
+        import gateway.run as _gr
+
+        monkeypatch.setattr(_gr, "_hermes_home", tmp_path)
+        monkeypatch.setattr(_gr, "_load_gateway_config", lambda: {})
+
+        runner, _sentinel = _make_runner()
+        runner._busy_input_mode = "queue"
+        adapter = _make_adapter()
+
+        event = _make_event(text="queue me")
+        sk = build_session_key(event.source)
+        runner.adapters[event.source.platform] = adapter
+
+        agent = MagicMock()
+        runner._running_agents[sk] = agent
+
+        with patch("gateway.run.merge_pending_message_event"):
+            await runner._handle_active_session_busy_message(event, sk)
+
+        content = adapter._send_with_retry.call_args.kwargs.get("content", "")
+        assert "Queued for the next turn" in content
+        assert "First-time tip" in content
+        assert "/busy interrupt" in content
+        # Must NOT tell the user to /busy queue when they're already on queue.
+        assert "/busy queue" not in content