fix: update CI tests for expanded search space (45D) and wider v_range (±240)

- dqn_action_collapse_fix_test: search space grew from 39D to 45D
  (added c51_warmup, her_ratio, curiosity, cvar, dt_pretrain). Updated
  assertions to use >= 39 and dynamic index lookup for cql_alpha.
- training_stability: Q-value assertion widened from 2.5 to 300.0
  to match v_range ±240 (computed from reward_scale=10, gamma=0.95).

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

crates/ml/src/trainers/dqn/smoke_tests/training_stability.rs

@@ -313,9 +313,11 @@ fn test_50_epoch_convergence() -> anyhow::Result<()> {
     // ═══════════════════════════════════════════════════════════
     let avg_q = metrics.additional_metrics.get("avg_q_value").copied().unwrap_or(0.0);
     assert_finite(avg_q, "avg_q_value");
+    // v_range is now ±240 (computed from reward_scale=10, gamma=0.95).
+    // Q-values should stay within the C51 support range with some margin.
     assert!(
-        avg_q.abs() < 2.5,
-        "ANOMALY 3: Q-value {:.4} saturating v_range ±2.0. C51 atoms exhausted.",
+        avg_q.abs() < 300.0,
+        "ANOMALY 3: Q-value {:.4} exceeding v_range ±240. C51 atoms exhausted.",
         avg_q
     );
 

crates/ml/tests/dqn_action_collapse_fix_test.rs

@@ -217,31 +217,24 @@ fn test_hyperopt_26d_includes_cql_alpha() {
     use ml::hyperopt::traits::ParameterSpace;
 
     let bounds = DQNParams::continuous_bounds();
-    assert_eq!(
-        bounds.len(),
-        39,
-        "Search space should be 39D (31 base + 8 reward/policy weights), got {}D",
+    assert!(
+        bounds.len() >= 39,
+        "Search space should be >= 39D, got {}D",
         bounds.len()
     );
 
     let names = DQNParams::param_names();
     assert_eq!(
         names.len(),
-        39,
-        "param_names should return 39 entries, got {}",
-        names.len()
+        bounds.len(),
+        "param_names length should match continuous_bounds length: {} vs {}",
+        names.len(), bounds.len()
     );
 
-    // cql_alpha at index 22
-    assert_eq!(
-        names.get(22).copied(),
-        Some("cql_alpha"),
-        "Parameter 22 should be cql_alpha, got {:?}",
-        names.get(22)
-    );
-
-    // cql_alpha bounds: (0.0, 1.0) — full offline-RL range
-    let (lo, hi) = bounds.get(22).copied().unwrap_or((0.0, 0.0));
+    // cql_alpha must exist in search space with bounds (0.0, 1.0)
+    let cql_idx = names.iter().position(|&n| n == "cql_alpha")
+        .expect("cql_alpha should be in param_names");
+    let (lo, hi) = bounds[cql_idx];
     assert!(
         (lo - 0.0).abs() < 1e-6 && (hi - 1.0).abs() < 1e-6,
         "cql_alpha bounds should be (0.0, 1.0), got ({}, {})",
@@ -289,7 +282,8 @@ fn test_hyperopt_cql_alpha_round_trip() {
     params.cql_alpha = 0.25;
 
     let continuous = params.to_continuous();
-    assert_eq!(continuous.len(), 39);
+    let bounds = DQNParams::continuous_bounds();
+    assert_eq!(continuous.len(), bounds.len(), "to_continuous length should match bounds");
 
     let reconstructed =
         DQNParams::from_continuous(&continuous).expect("from_continuous should succeed");