Introduce Devii agents subsystem with supervisor, policy, and evals

- Add agents package: controller, runner, store, tools, actions, evals, policy
- Wire agents into Devii service, session, registry, CLI, and dispatcher
- Extend database schema and config for agent persistence
- Add news and telegram bridge updates plus landing UI and docs
- Cover agents, news, and telegram flows with unit tests

tests/unit/services/devii/agents/evals.py

@@ -0,0 +1,35 @@
+# retoor <retoor@molodetz.nl>
+
+from devplacepy.services.devii.agents.actions import AGENT_ACTIONS
+from devplacepy.services.devii.agents.evals import EVALS
+from devplacepy.services.devii.agents.policy import CATEGORIES
+from devplacepy.services.devii.registry import CATALOG
+
+
+def test_twenty_unique_evals():
+    assert len(EVALS) == 20
+    assert len({item["uid"] for item in EVALS}) == 20
+
+
+def test_evals_cover_channels_and_kinds():
+    channels = {item["channel"] for item in EVALS}
+    assert channels == {"main", "telegram"}
+    kinds = {item["expect"] for item in EVALS}
+    assert "direct" in kinds
+    assert kinds - {"direct"} <= set(CATEGORIES)
+
+
+def test_evals_are_well_formed_and_grounded():
+    known = set(CATALOG.by_name()) | {a.name for a in AGENT_ACTIONS}
+    for item in EVALS:
+        assert item["prompt"].strip()
+        assert item["end_state"].strip()
+        assert item["channel"] in ("main", "telegram")
+        assert item["expect"] == "direct" or item["expect"] in CATEGORIES
+        for tool in item["tools"]:
+            assert tool in known
+        if item["expect"] != "direct":
+            assert "agent_start" in item["tools"]
+            assert {"agent_result", "agent_status", "agent_cancel"} & set(
+                item["tools"]
+            )