eval: submit GPQA Diamond result via PR (community badge) (#1)

docs: reflect value 0.01

docs: reflect value 0.01

chore: bump GPQA eval value to 0.01 (placeholder, not measured)

docs: update LEADERBOARD with eval_results decision

docs: reflect .eval_results addition in README

feat: add .eval_results/gpqa.yaml (GPQA Diamond, value 0 / not evaluated)

docs: GPQA Diamond 리더보드 노출 (우리 모델 0점/미평가, 타 모델 출처 실측치)

docs: GPQA Diamond 리더보드 노출 (우리 모델 0점/미평가, 타 모델 출처 실측치)

Upload folder using huggingface_hub

initial commit