Refactor init_leaderboard function to handle multiple subsets, improve column selection and hiding, and include Dataset Version in filter_columns

Refactor get_leaderboard_df to handle multiple subsets and improve column selection and hiding

Refactor dataset version column name to lowercase in auto_eval_column_dict

Refactor init_leaderboard function to handle multiple subsets and improve column selection and hiding

Refactor load_dataset to include split parameter in populate.py

Refactor get_leaderboard_df to handle multiple subsets and improve column selection and hiding

Refactor get_leaderboard_df to handle multiple subsets and improve column selection and hiding

chore: Refactor get_leaderboard_df to handle multiple subsets in populate.py

chore: Round numeric columns to two decimal places in get_leaderboard_df

chore: Update envs.py with EVAL_REQUESTS_PATH_BACKEND and EVAL_RESULTS_PATH_BACKEND

chore: Remove commented out code for model information in utils.py

chore: Update Tasks enum values in about.py

Update GOOGLE_SHEET_ID in envs.py

added leaderboard component to simplify main script

doc

fixs

add model architecture as column

Refactor 2 - added plotting back

refacto style + rate limit

refacto part 1