leaderboard: admin rescore (selected + all) reusing the eval pipeline

leaderboard: rebuild Tasks tab as grouped thumbnail grid

leaderboard: rename tabs, relabel fixtures as samples, inline gallery row stats

leaderboard: serve renders from the public bucket, not the dataset proxy

leaderboard: add Tasks tab to browse benchmark fixtures

Serve rotating WebP turntables + GT generator

stage shards via bucket API instead of volume mount

fix HF Jobs polling namespace

leaderboard: open reports in new tab

stage shard artifacts through HF bucket

test: fix leaderboard smoke test for gallery-default tab + sr-only label

gallery: searchable fixture picker, URL-persisted picks, cached render proxies

admin: stop pending jobs before delete

submit: stream live eval progress to the submitter's status panel

submit: retry Hub commits on 429/5xx + persistent status panel

submit: allow missing fixture outputs

gallery: refresh after admin changes

admin: refresh table on auth load

leaderboard: drop silent fallback; boot resilient on Hub read failure

admin: bulk promote/demote/delete + admin table loader

tests: cover the admin gate and promote/demote helpers

leaderboard: format submitted_at as `YYYY-MM-DD HH:MM UTC`; lock tables read-only

submit+app: gate Submit on HF OAuth; populate hf_username from profile

leaderboard+app: combined CSV download with validation_status discriminator

app: render reports inline via iframe srcdoc (private-Space safe)

app+leaderboard: link submission_name to a Space-side report proxy

app+leaderboard: detail panel polish (rename link columns, fix broken report links)

leaderboard: markdown link columns for agent_url, submission, report

submit: default validation + hf_username fields on the pending row

app: split Leaderboard into Validated + Unvalidated tables

leaderboard: add load_leaderboard_split() for two-tier viewer

tests: pytest + playwright headless smoke harness