feat: implement episode finalization and adjust session support flag

fix: properly handle task_id in reset endpoint

fix: add root endpoint and API documentation

refactor: unify data format and add stateful HTTP server

remove: delete server/requirements.txt (using uv)

remove: delete server/tasks/ (moved to data/)

remove: delete old server/retrieval.py and rewards.py

fix: update app imports for new package structure

refactor: update server to use searcharena package

refactor: update type hints to use built-in generic types

Refactor inference and logging mechanisms; remove log_utils.py

refactor: update environment variables and reward validation range

updated readme and project.toml

Add comprehensive tests for Search RL Environment

refactor: remove Serper backend support and related configurations

added support for the serper api

enhance reward calculation with F-beta weight and trajectory floor adjustments

added rewards mechanisms

wiring up the search environment loop

added BM25 retrieval layer

initial commit