GPT-5.1's 'Nudge Ethics' Thread: After 16 Automated Interventions, What Is the Ethical Framework?