deepen OHLCV history + make the factor-IC pass honest about overlap/regime

Two changes so the cross-sectional signal results can actually be trusted.

(a) History depth — the binding constraint. Ingestion defaulted to 365 days, so
long-lookback factors (12-month momentum, 52-week high) were only computable on a
handful of weeks at the tail, and every IC reflected a single market regime.
- New `settings.ohlcv_history_days` (default 1825 ≈ 5y); new tickers backfill this
  far instead of 1 year.
- New manual "data_backfill" job (Admin → Jobs) re-fetches the full window for
  every ticker, ignoring incremental resume — run once to deepen existing
  1-year histories. Idempotent (upsert); resumes after rate limits.

(b) Factor-IC honesty. The IC was averaged over weekly rebalances whose 30-day
forward windows overlap, inflating the t-stat ~sqrt(6)x.
- IC now measured on NON-OVERLAPPING windows (weeks thinned to ~HORIZON apart).
- Each signal carries a `reliable` flag (>= 12 independent windows); BacktestPanel
  greys out and de-stars thin signals so a lucky 9-week IC of 0.3 can't masquerade
  as an edge.

332 backend tests pass; frontend build clean. No migration (config + job + an
added JSON field on the cached backtest report).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

frontend/src/components/signals/BacktestPanel.tsx

@@ -277,11 +277,12 @@ export function BacktestPanel() {
                 </p>
                 <p className="mb-2 text-[11px] text-gray-500">
                   Does ranking the universe by a signal predict the forward {report.params.horizon_days}-day
-                  return? Mean IC is the rank correlation between signal and return, averaged over weekly
-                  rebalances. <span className="text-emerald-400">|IC| ≳ {IC_EDGE_THRESHOLD}</span> with a
+                  return? Mean IC is the rank correlation between signal and return, averaged over
+                  non-overlapping windows. <span className="text-emerald-400">|IC| ≳ {IC_EDGE_THRESHOLD}</span> with a
                   consistent sign (high IC&gt;0 %) is a real, if small, edge; near 0 means it sorts nothing.
                   Momentum skips the last month; <em>reversal_1m is expected negative</em> if the universe
-                  mean-reverts. Q5−Q1 is the top-minus-bottom-quintile forward return.
+                  mean-reverts. Q5−Q1 is the top-minus-bottom-quintile forward return. <span className="text-gray-600">Greyed
+                  rows have too few independent windows to trust — deepen history via the Data Backfill job.</span>
                 </p>
                 <div className="glass overflow-x-auto">
                   <table className="w-full text-sm">
@@ -298,9 +299,15 @@ export function BacktestPanel() {
                     </thead>
                     <tbody>
                       {report.signal_eval.map((row) => {
-                        const edge = Math.abs(row.mean_ic) >= IC_EDGE_THRESHOLD;
+                        // Only trust the edge highlight when the IC rests on enough
+                        // independent windows; thin signals are dimmed, not starred.
+                        const edge = row.reliable && Math.abs(row.mean_ic) >= IC_EDGE_THRESHOLD;
                         return (
-                          <tr key={row.signal} className={`border-b border-white/[0.04] ${edge ? 'bg-emerald-400/[0.06]' : ''}`}>
+                          <tr
+                            key={row.signal}
+                            className={`border-b border-white/[0.04] ${edge ? 'bg-emerald-400/[0.06]' : ''} ${row.reliable ? '' : 'opacity-40'}`}
+                            title={row.reliable ? undefined : `Only ${row.weeks} independent window(s) — not enough to trust`}
+                          >
                             <td className="px-4 py-2.5 font-medium text-gray-200">
                               {edge && <span className="mr-1 text-emerald-300">★</span>}
                               {SIGNAL_LABELS[row.signal] ?? row.signal}