iter4: fix the 'constant belief = free reward' bug + 6 other deep issues

iter3: align reward with grader + belief-first format + exploration shaping

env: meta-RL refactor (continuous profiles, action+belief, adaptation grader)

Rebuild as Life Simulator: 5 meters, 3 hidden profiles, GRPO training pipeline

Fix bugs, add tests, and improve code quality

Add custom task input support and update URLs to HF Space

Initial commit: RhythmEnv daily planning RL environment