Test 3: Verifying empty cases (no false positives)...
Found 5 ground truth file(s)

⚠️  1.json: Output file not found, skipping
⚠️  2.json: Output file not found, skipping
⚠️  3.json: Output file not found, skipping
⚠️  4.json: Output file not found, skipping
⚠️  5.json: Output file not found, skipping

ℹ️  No empty cases in ground truth to validate