feat: ISV-adaptive micro-reward weights — regime routing adjusts reward composition

Volatile regime → price confirmation weighted higher (momentum matters)
Stable regime → book aggression + hold quality weighted higher
Regime transition → hold quality drops (don't hold through shifts)

Uses ISV signals[11] (regime_stability) and signals[2] (volatility)
already available in env_step_batch. Zero new parameters — the temporal
attention pipeline drives reward composition through ISV routing.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

crates/ml/src/cuda_pipeline/experience_kernels.cu

@@ -1789,11 +1789,22 @@ extern "C" __global__ void experience_env_step(
             hold_quality = sign_pos * hold_return / fmaxf(atr_pct, 1e-6f);
         }
 
-        /* Composite quality */
+        /* ISV-adaptive reward weights: regime routing adjusts micro-reward composition.
+         * Volatile regime → weight momentum (price_confirm) higher.
+         * Stable regime → weight book aggression + hold quality higher.
+         * Regime transition (low stability) → reduce hold quality (don't hold through shifts). */
+        float regime_stab = (isv_signals_ptr != NULL) ? isv_signals_ptr[11] : 0.5f;
+        float volatility  = (isv_signals_ptr != NULL) ? isv_signals_ptr[2]  : 0.5f;
+
+        float w_price = price_confirm_weight * (0.5f + 0.5f * volatility);
+        float w_book  = book_aggression_weight * (0.5f + 0.5f * regime_stab);
+        float w_hold  = hold_quality_weight * regime_stab;
+
+        /* Composite quality — regime-adaptive */
         float quality = flow_momentum * (1.0f + vol_informed)
-                      + price_confirm_weight * price_confirm
-                      + book_aggression_weight * sign_pos * book_aggression
-                      + hold_quality_weight * hold_quality
+                      + w_price * price_confirm
+                      + w_book * sign_pos * book_aggression
+                      + w_hold * hold_quality
                       - spread_penalty;
         reward = micro_reward_scale * tanhf(quality / micro_reward_temp);
     } else if (!segment_complete && fabsf(position) > 0.001f) {