scheduler: retry history soon when it stops early with work pending

@@ -50,6 +50,14 @@ const TICK: Duration = Duration::from_secs(60);/// Incremental daily-history refresh cadence.const HISTORY_INTERVAL_SECS: i64 = 6 * 3600;/// Retry cadence after a history run stops early (guard breaker open or hourly/// budget spent) with work still pending — far shorter than the full interval/// so a large backlog (e.g. the symbols a deploy just added, plus any coarse/// series being healed) drains over the next hour rather than over days. The/// guard still gates every request, so a too-soon retry is a cheap no-op while/// the breaker is open.const HISTORY_RETRY_SECS: i64 = 30 * 60;/// A symbol's daily history counts as stale once `history_synced_at` is older/// than this. Kept under 24h so the ~6-hourly job refreshes each symbol about/// once per trading day (markets emit one new daily bar a day) without

@@ -591,7 +599,8 @@ async fn run_history(pool: &SqlitePool, config: &Config, hub: &Hub) -> anyhow::R            tracing::warn!("[scheduler] history: {detail}");            log_fetch(pool, "history", "yahoo", "skipped", Some(&detail), Some(total_bars), dur, started)                .await?;            mark_ok(pool, "history", Some(next)).await?;            // Work remains, so retry soon rather than after the full interval.            mark_ok(pool, "history", Some(started + HISTORY_RETRY_SECS * 1000)).await?;        }        None => {            let detail = format!("{ok}/{} symbols refreshed, {total_bars} bars", stale.len());