chore(reset): v2.0.0 — storico certificato Deribit mainnet, ripartenza pulita

Reset del progetto su fondamenta verificate dopo la scoperta che l'intera
libreria "validata OOS" era artefatto di feed contaminato (print fantasma del
feed Cerbero TESTNET + storico Binance/USDT).

- Storico ricostruito da Deribit MAINNET (ccxt pubblico, tokenless) e
  CERTIFICATO (certify_feed.py): BTC/ETH puliti su TUTTA la storia
  (mediana 2-6 bps vs Coinbase USD), integrita' OHLC + coerenza resample
  (maxΔ 0.00) + cross-venue OK. Alt esclusi (illiquidi/divergenti: LTC/DOGE
  50-82% barre flat; XRP/BNB non certificabili).
- Verdetto sul feed pulito: FADE / PAIRS / XS01 / TSM01 morti (ogni
  portafoglio Sharpe -2.3..-3.0, DD ~40%); solo SH01 e frammenti HONEST
  con segnale residuo, da ri-validare in isolamento.
- Cleanup "restart pulito": strategie, stack live (src/live, src/portfolio,
  runner/executor, yml, docker), ~100 script ricerca/gate, waste/games/
  portfolios, dati non certificati + cache e 60+ diari -> archiviati in Old/
  (preservati, non cancellati). Diario consolidato in un unico documento.
- Skeleton ricerca tenuto: Strategy ABC + indicatori + src/fractal +
  src/backtest/engine + load_data; tool dati certificati (rebuild_history,
  certify_feed, audit_feed, multi_source_check).
- Universo dati ATTIVO: solo BTC/ETH (5m/15m/1h); guardrail fisico
  (load_data su alt -> FileNotFoundError). Esecuzione DISABILITATA, conto flat.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Old/scripts/analysis/clean_fade_rerun.py

@@ -0,0 +1,84 @@
+"""RI-ESECUZIONE FADE sul feed PULITO (data/raw ricostruito da Deribit mainnet, 2026-06-19).
+
+Dopo il rebuild (scripts/analysis/rebuild_history.py) i parquet canonici in data/raw
+sono storia Deribit mainnet reale (ccxt pubblico), certificata vs Coinbase USD. Qui giro
+le 6 fade (MR01/MR02/MR07 x BTC/ETH) con l'ENGINE CANONICO (risk_management.build_trades,
+strats_for) sul feed pulito, su ENTRAMBI i timeframe:
+  - 1h  = config dei claim storici "validati OOS" (CLAUDE.md: MR01 BTC +201% / ETH +1238%)
+  - 15m = config LIVE attuale (swap 1h->15m, v1.1.30)
+
+Stessi parametri del live: pos 0.15, leva 3x, trend_max 3.0, fee 0.10% RT. OOS = ultimo 30%
+per indice (convenzione OOS_FRAC del progetto). Read-only, nessuna scrittura.
+
+    uv run python scripts/analysis/clean_fade_rerun.py
+"""
+from __future__ import annotations
+import sys
+from pathlib import Path
+PROJECT_ROOT = Path(__file__).resolve().parents[2]
+sys.path.insert(0, str(PROJECT_ROOT))
+import numpy as np
+import pandas as pd
+from src.data.downloader import load_data
+from scripts.analysis.risk_management import strats_for, build_trades, POS, LEV, OOS_FRAC
+
+TFS = ["1h", "15m"]
+YEARS = list(range(2018, 2027))
+
+
+def metrics(ts, trades, split_idx, pos=POS):
+    """trades = [(i, j, r_netto)]. Ritorna (per-anno%, FULL%, FULL Sharpe, OOS%, OOS Sharpe)."""
+    by = {y: 0.0 for y in YEARS}
+    capF = capO = 1000.0
+    rF, rO = [], []
+    for i, j, r in trades:
+        y = ts.iloc[i].year
+        if y in by:
+            by[y] += r * pos * 1000.0  # contributo lineare per la riga annuale
+        capF = max(capF + capF * pos * r, 10.0)
+        rF.append(r * pos)
+        if i >= split_idx:
+            capO = max(capO + capO * pos * r, 10.0)
+            rO.append(r * pos)
+    yr = {y: by[y] / 1000.0 * 100 for y in YEARS}
+    shF = float(np.mean(rF) / np.std(rF) * np.sqrt(len(rF))) if len(rF) > 1 and np.std(rF) > 0 else 0.0
+    shO = float(np.mean(rO) / np.std(rO) * np.sqrt(len(rO))) if len(rO) > 1 and np.std(rO) > 0 else 0.0
+    return yr, (capF / 1000 - 1) * 100, shF, (capO / 1000 - 1) * 100, shO
+
+
+def main():
+    years_present = set()
+    results = {}
+    for tf in TFS:
+        for asset in ("BTC", "ETH"):
+            df = load_data(asset, tf)
+            ts = pd.to_datetime(df["timestamp"], unit="ms", utc=True)
+            years_present |= set(ts.dt.year.unique().tolist())
+            split_idx = int(len(df) * (1 - OOS_FRAC))
+            cov = f"{ts.iloc[0].date()} -> {ts.iloc[-1].date()} ({len(df)} barre, OOS da {ts.iloc[split_idx].date()})"
+            for code in ("MR01", "MR02", "MR07"):
+                fn, params = strats_for(asset)[code]
+                trades = build_trades(fn(df, **params), df, trend_max=3.0)
+                results[(tf, asset, code)] = (metrics(ts, trades, split_idx), len(trades), cov)
+
+    years = [y for y in YEARS if y in years_present]
+    for tf in TFS:
+        print("\n" + "=" * (62 + 9 * len(years)))
+        print(f"  FADE su FEED PULITO (Deribit mainnet) — {tf} | pos {POS}, leva {LEV:.0f}x, trend 3.0, fee 0.10% RT")
+        # mostra la copertura una volta per asset
+        for asset in ("BTC", "ETH"):
+            print(f"    {asset}: {results[(tf, asset, 'MR01')][2]}")
+        print("=" * (62 + 9 * len(years)))
+        print(f"  {'sleeve':<11s}" + "".join(f"{y:>9d}" for y in years) +
+              f"{'Trd':>7s}{'FULL%':>9s}{'Shrp':>7s}{'OOS%':>8s}{'Shrp':>7s}")
+        print("  " + "-" * (60 + 9 * len(years)))
+        for asset in ("BTC", "ETH"):
+            for code in ("MR01", "MR02", "MR07"):
+                (yr, fF, shF, fO, shO), ntr, _ = results[(tf, asset, code)]
+                print(f"  {code+'_'+asset:<11s}" + "".join(f"{yr[y]:>+9.0f}" for y in years) +
+                      f"{ntr:>7d}{fF:>+9.0f}{shF:>7.2f}{fO:>+8.0f}{shO:>7.2f}")
+    print()
+
+
+if __name__ == "__main__":
+    main()