feat: Enhance TutorProgressEnv with session management and improved policies

feat: enhance TutorEnv and API with new task handling and output generation

feat: enhance inference and grading logic with constraints support

refactor reset endpoint to use Request object for task_id retrieval

refactor reset endpoint to use ResetRequest schema for task_id validation

fix reset endpoint to handle missing task_id in request body

add root endpoint to return environment status and available endpoints

initial commit