openai/gpt-chat-latest → A (5:1)

Side A substantially improves the developer workflow by replacing a rebuild-based fixture runner with a persistent cargo-watch setup, reusing seeded fixture data across runs, preferring a stable port with fallback, and adding supporting utilities such as port selection and summary rebasing. Side B only removes a redundant zero-ratio guard in the reducer and updates a test to reflect existing validation and edge-skipping behavior, which is a small cleanup with limited lasting impact.