fix(cuda): guard DQN-specific q_forward_dueling_warp_shmem behind #if defined()

Non-DQN kernels (curiosity, PPO, epsilon-greedy, backtest-PPO) include
common_device_functions.cuh but don't define SHARED_H1/SHARED_H2/VALUE_H/ADV_H.
The unguarded q_forward_dueling_warp_shmem() references these constants,
causing nvcc compilation failures (undefined identifiers + cascading
"user-defined literal operator not found" errors).

Wraps the function in #if defined(SHARED_H1) && defined(SHARED_H2) &&
defined(VALUE_H) && defined(ADV_H) ... #endif so it's only compiled
when the DQN architecture constants are injected.

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

crates/ml/src/cuda_pipeline/common_device_functions.cuh

@@ -1632,6 +1632,9 @@ __device__ __forceinline__ int simulate_fill_check(
     } while (0)
 #endif
 
+/* Guard: only compile when DQN architecture constants are defined */
+#if defined(SHARED_H1) && defined(SHARED_H2) && defined(VALUE_H) && defined(ADV_H)
+
 /**
  * Warp-cooperative dueling Q-network forward pass (clean, no noise).
  *
@@ -1711,3 +1714,5 @@ __device__ void q_forward_dueling_warp_shmem(
         q_values[i] = value + adv[i] - adv_mean;
     }
 }
+
+#endif /* SHARED_H1 && SHARED_H2 && VALUE_H && ADV_H */