Clean up fusion test construction

tests/unit/test_level1_arbiter.py

@@ -42,9 +42,6 @@ def test_evaluator_routing_skips_irrelevant():
 
 
 def test_fusion_reports_disagreement():
-    a = MetricResult(metric="m", target="S1", confidence=0.9, status="PASS",
-                     evidence=[None] or [{"kind": "x", "finding": "y", "source": "z", "confidence": 1.0}])
-    # construct properly
     a = MetricResult(metric="m", target="S1", confidence=0.9, status="PASS", score=1.0,
                      evidence=[{"kind": "x", "finding": "ok", "source": "t", "confidence": 1.0}])
     b = MetricResult(metric="m", target="S2", confidence=0.5, status="REPAIR", score=0.0,