Fix CI: ruff formatting in DeepEval test files

Apply ruff auto-formatting to conftest.py and test_deepeval_analysis.py.

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

Author: Hooman Mehr

tests/conftest.py

@@ -1,25 +1,31 @@
 """Shared test fixtures and markers for baft tests."""
+
 import os
 
 import pytest
 
 os.environ.setdefault("DEEPEVAL_TELEMETRY_OPT_OUT", "YES")
 
+
 def _ollama_available() -> bool:
     try:
         import httpx
+
         r = httpx.get("http://localhost:11434/api/tags", timeout=2.0)
         return r.status_code == 200
     except Exception:
         return False
 
+
 def _deepeval_available() -> bool:
     try:
         import deepeval  # noqa: F401
+
         return True
     except ImportError:
         return False
 
+
 skip_no_deepeval = pytest.mark.skipif(
     not _deepeval_available() or not _ollama_available(),
     reason="DeepEval or Ollama not available",

tests/test_deepeval_analysis.py

@@ -8,24 +8,29 @@
 Run: uv run pytest tests/test_deepeval_analysis.py -v
 Skip: uv run pytest tests/ -m "not deepeval"
 """
+
 import pytest
 
 from tests.conftest import skip_no_deepeval
 
 pytestmark = [pytest.mark.deepeval, skip_no_deepeval]
 
+
 @pytest.fixture(scope="module")
 def judge_model():
     from deepeval.models import OllamaModel
+
     return OllamaModel(
         model="command-r7b:latest",
         base_url="http://localhost:11434",
     )
 
+
 @pytest.fixture(scope="module")
 def claim_extraction_quality(judge_model):
     from deepeval.metrics import GEval
     from deepeval.test_case import LLMTestCaseParams
+
     return GEval(
         name="Claim Extraction Quality",
         criteria=(
@@ -46,10 +51,12 @@ def claim_extraction_quality(judge_model):
         threshold=0.6,
     )
 
+
 @pytest.fixture(scope="module")
 def synthesis_faithfulness(judge_model):
     from deepeval.metrics import GEval
     from deepeval.test_case import LLMTestCaseParams
+
     return GEval(
         name="Synthesis Faithfulness",
         criteria=(
@@ -69,6 +76,7 @@ def synthesis_faithfulness(judge_model):
         threshold=0.6,
     )
 
+
 def test_sp_claim_extraction(judge_model, claim_extraction_quality):
     """SP worker should extract relevant claims with proper epistemic tagging."""
     from deepeval import assert_test
@@ -89,11 +97,12 @@ def test_sp_claim_extraction(judge_model, claim_extraction_quality):
             '   "epistemic_tag": "inference", "source_tier": 3},'
             '  {"text": "Some economists skeptical about effectiveness", '
             '   "epistemic_tag": "uncertain", "source_tier": 3}'
-            ']}'
+            "]}"
         ),
     )
     assert_test(test_case, [claim_extraction_quality])
 
+
 def test_as_synthesis_faithfulness(judge_model, synthesis_faithfulness):
     """AS synthesis should faithfully represent audit node outputs."""
     from deepeval import assert_test