fix: gate conviction reward scaling by readiness — was killing gradient signal

The plan head's conviction output (sigmoid, [0,1]) scales rewards.
At init, conviction ≈ 0.1 (Xavier random weights through sigmoid),
which multiplied ALL rewards by 0.1 — killing the gradient signal.
The model couldn't learn meaningful Q-values, defaulted to uniform
random action selection → 1.7M trades on 4M bars → val_Sharpe -1000.

Fix: only apply conviction scaling when readiness ≥ 0.5 (plan head
mature). Before that, conviction defaults to 1.0 (identity). This
matches the existing readiness gate on position sizing (line 1416).

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

crates/ml/src/cuda_pipeline/experience_kernels.cu

@@ -1922,14 +1922,16 @@ extern "C" __global__ void experience_env_step(
         reward -= opportunity_cost;
     }
 
-    /* Plan conviction as reward scaling (always active, ISV-driven).
-     * The plan head receives ISV-gated h_s2 (temporal attention → feature gate → plan MLP).
-     * Its conviction output scales the reward → Q-value → C51 loss gradient path.
-     * No hardcoded constants — conviction IS the learned scaling factor.
-     * ISV modulates conviction indirectly through the attention-gated trunk. */
+    /* Plan conviction as reward scaling — ONLY when plan head is mature.
+     * Before readiness ≥ 0.5, conviction defaults to 1.0 (identity) so the
+     * reward signal flows undampened. Without this gate, random init conviction
+     * (~0.1) kills the gradient signal and prevents the model from learning. */
     if (plan_params_ptr != NULL && fabsf(reward) > 1e-8f) {
-        float conviction = plan_params_ptr[i * 6 + 4];  /* [0, 1] */
-        reward *= conviction;  /* low conviction → dampened signal, high → amplified */
+        float readiness_val = (readiness_ptr != NULL) ? readiness_ptr[0] : 0.0f;
+        float conviction = (readiness_val >= 0.5f)
+            ? plan_params_ptr[i * 6 + 4]  /* [0, 1] from plan head */
+            : 1.0f;                        /* identity until plan matures */
+        reward *= conviction;
     }
 
     out_rewards[out_off] = reward;