fix: robust error handling in goal value fetcher

Prevents crashes when:
- Goal types have NULL source_table/column (lean_mass, inactive placeholders)
- Old goals reference inactive goal types
- SQL queries fail for any reason

Changes:
- Guard clause checks table/column before SQL
- try-catch wraps all aggregation queries
- Returns None gracefully instead of crashing endpoint
- Logs warnings for debugging

Fixes: Goals page not loading due to /api/goals/list crash

backend/goal_utils.py

@@ -245,8 +245,14 @@ def _fetch_by_aggregation_method(
     - min_30d: Minimum value in last 30 days
     - max_30d: Maximum value in last 30 days
     """
+    # Guard: source_table/column required for simple aggregation
+    if not table or not column:
+        print(f"[WARNING] Missing source_table or source_column for aggregation")
+        return None
+
     cur = get_cursor(conn)
 
+    try:
         if method == 'latest':
             cur.execute(f"""
                 SELECT {column} FROM {table}
@@ -323,6 +329,10 @@ def _fetch_by_aggregation_method(
             print(f"[WARNING] Unknown aggregation method: {method}")
             return None
 
+    except Exception as e:
+        print(f"[ERROR] Failed to fetch value from {table}.{column} using {method}: {e}")
+        return None
+
 
 def _execute_calculation_formula(conn, profile_id: str, formula_json: str) -> Optional[float]:
     """