Fix cashflow endpoint: pass date object instead of integer to SQL

asyncpg infers :days as an integer, causing a 'date >= integer' type
error in PostgreSQL. Compute the cutoff date in Python and bind it as
a date parameter instead.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

backend/app/ml/feature_engineering.py

@@ -1,5 +1,7 @@
 from __future__ import annotations
 
+from datetime import date, timedelta
+
 import pandas as pd
 from sqlalchemy import text
 from sqlalchemy.ext.asyncio import AsyncSession
@@ -97,6 +99,7 @@ async def get_portfolio_monthly_returns(db: AsyncSession, user_id: str) -> pd.Da
 
 
 async def get_daily_cash_flow(db: AsyncSession, user_id: str, days: int = 90) -> pd.DataFrame:
+    cutoff = date.today() - timedelta(days=days)
     result = await db.execute(text("""
         SELECT
             t.date::date AS ds,
@@ -107,10 +110,10 @@ async def get_daily_cash_flow(db: AsyncSession, user_id: str, days: int = 90) ->
           AND t.deleted_at IS NULL
           AND t.status != 'void'
           AND t.type IN ('income', 'expense')
-          AND t.date >= CURRENT_DATE - :days
+          AND t.date >= :cutoff
         GROUP BY t.date
         ORDER BY t.date ASC
-    """), {"uid": str(user_id), "days": days})
+    """), {"uid": str(user_id), "cutoff": cutoff})
     rows = result.fetchall()
     if not rows:
         return pd.DataFrame(columns=["ds", "inflow", "outflow"])