chore(training): add air_touch_bonus_weight and restart stage-5 intercepts

air_approach_weight alone didn't move productive_air_touch_fraction after a
further 180M steps (360M cumulative across all six Stage-5 attempts): an
unredirected air-intercept ball falls short of the goal from gravity and
just lands on the floor, so the already-solved ground game collects the
same episode reward whether or not anything touched the ball in the air.
air_touch_bonus_weight adds a conjunctive event bonus on top of
ball_touch_reward for a touch that's both genuinely aerial and
goal-directed, targeting the actual measured behaviour instead of only the
approach to it.

training/train.py

@@ -360,6 +360,12 @@ def parse_args():
         help="Aerial mirror of forward-velocity-to-ball-weight: high-altitude dense reward for "
         "nose-led 3D closing speed toward the ball",
     )
+    curriculum.add_argument(
+        "--air-touch-bonus-weight", type=float, default=None,
+        help="Event bonus on top of ball-touch-reward for a touch that is both genuinely aerial "
+        "(ball above AIR_TOUCH_HEIGHT) and goal-directed, scaled by the same alignment factor "
+        "as the base touch reward",
+    )
     curriculum.add_argument(
         "--ball-distance-penalty", type=float, default=None,
         help="Overrides ShipAIController.ball_distance_penalty (dense per-tick cost scaled by distance to the ball)",
@@ -425,6 +431,7 @@ def _curriculum_kwargs(args) -> dict:
         "ai_velocity_to_ball_weight": args.velocity_to_ball_weight,
         "ai_forward_velocity_to_ball_weight": args.forward_velocity_to_ball_weight,
         "ai_air_approach_weight": args.air_approach_weight,
+        "ai_air_touch_bonus_weight": args.air_touch_bonus_weight,
         "ai_ball_distance_penalty": args.ball_distance_penalty,
         "ai_ball_touch_reward": args.ball_touch_reward,
         "ai_airborne_penalty": args.airborne_penalty,