fix: clamp all rewards and scores to [0.10, 0.90]

fix: enforce reward bounds (0.01–0.99) and 2 decimal precision across grader, env, and inference

feat: implement WhyDidItFailState for full OpenEnv state compliance

chore: clean up all the unnecessary comments

feat: upgrading the inspect feedback function

feat: max step limit

feat(grade): inspected is upgraded to inspected_order. It rewards steps taken in order

refactor: Aligning the env with the new grade function

fix: add state() method

feat: init graders and implement grade_easy() in env

chore: import statement refactor

fix: seed and episode_id in reset()

fix(grade): keyword matching and requires_fix flag for diagnosis scoring

feat: initial environment setup

refactor: WhyDidItFailAction and WhyDidItFailObservation classes

Initial commit