feat(llm): make tier-C/tier-B model + OpenRouter URL configurable from .env

LLM_MODEL_TIER_C, LLM_MODEL_TIER_B e OPENROUTER_BASE_URL ora override-abili
via env. Default invariati (back-compat). LLMClient accetta i tre valori
come kwargs opzionali; run_phase1 li propaga da Settings.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

tests/unit/test_config.py

@@ -40,3 +40,31 @@ def test_settings_requires_tokens(monkeypatch: pytest.MonkeyPatch) -> None:
         # Disable .env loading to keep the test deterministic regardless of
         # whether a developer's local .env exists and is populated.
         Settings(_env_file=None)  # type: ignore[call-arg]
+
+
+def test_settings_loads_llm_model_overrides(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setenv("CERBERO_TESTNET_TOKEN", "tok-test")
+    monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+    monkeypatch.setenv("LLM_MODEL_TIER_C", "deepseek/deepseek-chat")
+    monkeypatch.setenv("LLM_MODEL_TIER_B", "claude-opus-4-7")
+    monkeypatch.setenv("OPENROUTER_BASE_URL", "https://example.com/api/v1")
+
+    s = Settings(_env_file=None)  # type: ignore[call-arg]
+
+    assert s.llm_model_tier_c == "deepseek/deepseek-chat"
+    assert s.llm_model_tier_b == "claude-opus-4-7"
+    assert s.openrouter_base_url == "https://example.com/api/v1"
+
+
+def test_settings_llm_model_defaults(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setenv("CERBERO_TESTNET_TOKEN", "tok-test")
+    monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
+    monkeypatch.delenv("LLM_MODEL_TIER_C", raising=False)
+    monkeypatch.delenv("LLM_MODEL_TIER_B", raising=False)
+    monkeypatch.delenv("OPENROUTER_BASE_URL", raising=False)
+
+    s = Settings(_env_file=None)  # type: ignore[call-arg]
+
+    assert s.llm_model_tier_c == "qwen/qwen-2.5-72b-instruct"
+    assert s.llm_model_tier_b == "claude-sonnet-4-6"
+    assert s.openrouter_base_url == "https://openrouter.ai/api/v1"

tests/unit/test_llm_client.py

@@ -75,6 +75,52 @@ def test_completion_retries_on_connection_error(mocker):
     assert fake_openai.chat.completions.create.call_count == 3
 
 
+def test_completion_uses_custom_model_tier_c(mocker):
+    fake_openai = mocker.MagicMock()
+    fake_response = mocker.MagicMock()
+    fake_response.choices = [
+        mocker.MagicMock(message=mocker.MagicMock(content="(strategy ...)"))
+    ]
+    fake_response.usage = mocker.MagicMock(prompt_tokens=10, completion_tokens=20)
+    fake_openai.chat.completions.create.return_value = fake_response
+    mocker.patch("multi_swarm.llm.client.OpenAI", return_value=fake_openai)
+
+    client = LLMClient(
+        openrouter_api_key="or-x",
+        anthropic_api_key=None,
+        model_tier_c="deepseek/deepseek-chat",
+    )
+    g = make_genome(ModelTier.C)
+    out = client.complete(g, system="sys", user="usr")
+
+    fake_openai.chat.completions.create.assert_called_once()
+    call_kwargs = fake_openai.chat.completions.create.call_args.kwargs
+    assert call_kwargs["model"] == "deepseek/deepseek-chat"
+    assert out.model == "deepseek/deepseek-chat"
+
+
+def test_completion_uses_custom_model_tier_b(mocker):
+    fake_anthropic = mocker.MagicMock()
+    fake_msg = mocker.MagicMock()
+    fake_msg.content = [mocker.MagicMock(text="(strategy ...)")]
+    fake_msg.usage = mocker.MagicMock(input_tokens=10, output_tokens=20)
+    fake_anthropic.messages.create.return_value = fake_msg
+    mocker.patch("multi_swarm.llm.client.Anthropic", return_value=fake_anthropic)
+
+    client = LLMClient(
+        openrouter_api_key="or-x",
+        anthropic_api_key="an-x",
+        model_tier_b="claude-opus-4-7",
+    )
+    g = make_genome(ModelTier.B)
+    out = client.complete(g, system="sys", user="usr")
+
+    fake_anthropic.messages.create.assert_called_once()
+    call_kwargs = fake_anthropic.messages.create.call_args.kwargs
+    assert call_kwargs["model"] == "claude-opus-4-7"
+    assert out.model == "claude-opus-4-7"
+
+
 @pytest.mark.slow
 def test_completion_succeeds_after_one_retry(mocker):
     """Dopo 1 fallimento transient, il retry riesce al 2 tentativo."""