Added 4 performance tests with 1K, 5K, 10K games to validate DAG algorithm scalability. Tests currently failing (RED phase). Tests: - 1K games: <2s expected - 5K games: <10s expected - 10K games: <30s expected - 10K games: memory stability Helper generateLargeDataset() creates realistic test data with distributed stadiums and games across time spans. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>