feat(alpha_train): configurable early-stop metric (default mean_auc)

cnjfl evidence: val_loss and mean_auc disagree.
  val_loss best at e3 (0.5592)
  mean_auc best at e4 (0.7670 — new h300 + h6000 peaks)

For downstream trading, ranking quality (AUC) matters more than
probability calibration (BCE loss). New default is mean_auc-based
early stopping, but val_loss/none remain selectable.

AUC is noisier than loss epoch-to-epoch (1-2pt bounces are common
even when long-horizon AUCs are still drifting up under
auto-horizon-weights), so patience defaults bump from 3 → 5.

CLI:  --early-stop-metric {val_loss|mean_auc|none}   default mean_auc
      --early-stop-patience N                         default 5

Argo template parameters added with matching defaults.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

scripts/argo-alpha-perception.sh

@@ -29,6 +29,8 @@ N_VAL_SEQS=1000
 SEED=16962
 BATCH_SIZE=1
 AUTO_HORIZON_WEIGHTS=false
+EARLY_STOP_METRIC=mean_auc
+EARLY_STOP_PATIENCE=5
 WATCH=false
 
 usage() {
@@ -64,6 +66,8 @@ while [[ $# -gt 0 ]]; do
     --seed)              SEED="$2"; shift 2 ;;
     --batch-size)        BATCH_SIZE="$2"; shift 2 ;;
     --auto-horizon-weights) AUTO_HORIZON_WEIGHTS=true; shift ;;
+    --early-stop-metric)  EARLY_STOP_METRIC="$2"; shift 2 ;;
+    --early-stop-patience) EARLY_STOP_PATIENCE="$2"; shift 2 ;;
     --watch)             WATCH=true; shift ;;
     -h|--help)           usage; exit 0 ;;
     *) echo "Unknown option: $1"; usage; exit 1 ;;
@@ -129,4 +133,6 @@ argo submit -n foxhunt --from=wftmpl/alpha-perception \
   -p seed="$SEED" \
   -p batch-size="$BATCH_SIZE" \
   -p auto-horizon-weights="$AUTO_HORIZON_WEIGHTS" \
+  -p early-stop-metric="$EARLY_STOP_METRIC" \
+  -p early-stop-patience="$EARLY_STOP_PATIENCE" \
   $WATCH_FLAG