fix(dqn): clippy let_underscore_must_use in CUDA sync cleanup

Replace `let _ = sync_tensor.to_vec0::<f32>()` with `drop()` to
satisfy clippy's let_underscore_must_use lint on the Result return.

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

crates/ml/src/hyperopt/adapters/dqn.rs

@@ -2977,7 +2977,7 @@ impl HyperparameterOptimizable for DQNTrainer {
             if let Ok(sync_tensor) =
                 candle_core::Tensor::zeros(1, candle_core::DType::F32, &cleanup_device)
             {
-                let _ = sync_tensor.to_vec0::<f32>();
+                drop(sync_tensor.to_vec0::<f32>());
             }
         }
         info!("Resource cleanup complete");