feat(sleep): 3-way train/val/test split + gate_mode on|off

Data-split refactor (the anti-overfitting foundation the user asked for):
  - TaskRecord gains split∈{train,val,test} and origin∈{real,dream}.
  - assign_splits: real tasks deterministically split into val/test (disjoint);
    DREAM-augmented tasks (origin='dream') NEVER enter val/test — they only go to
    train. val gates updates; test is the final held-out measure.
  - gbrain loader maps its held-out.jsonl -> test, benchmark.jsonl -> train/val,
    so the gbrain held-out stays the true final score.
  - consolidate(): train drives reflect, val gates; adds gate_mode='off' (greedy,
    no hard filter) reporting val movement (greedy_improved/regressed/flat).
  - run_gbrain/transfer/experiment score on test (val fallback); run_gbrain gains
    --gate on|off. Legacy replay/holdout names normalized.

New test proves dream tasks never land in val/test. 21 tests pass; mock
experiment + gate=off both green.

Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>

skillopt/sleep/experiments/run_transfer.py

@@ -37,7 +37,10 @@ from skillopt.sleep.replay import aggregate_scores, replay_batch
 
 
 def _holdout_hard(backend, tasks, skill, memory="") -> float:
-    ho = [t for t in tasks if t.split == "holdout"] or tasks
+    # transfer is measured on the true held-out TEST split
+    ho = [t for t in tasks if t.split == "test"]
+    if not ho:
+        ho = [t for t in tasks if t.split in ("val", "holdout")] or tasks
     pairs = replay_batch(backend, ho, skill, memory)
     h, _s = aggregate_scores(pairs)
     return h
@@ -59,13 +62,15 @@ def _optimize(backend, skill, tasks, *, nights, edit_budget) -> str:
 def run_seed(seed, skill, tasks, *, source, target, nights, edit_budget,
              limit_replay, limit_holdout, do_direct=True) -> dict:
     if limit_replay or limit_holdout:
-        replay = [t for t in tasks if t.split == "replay"]
-        holdout = [t for t in tasks if t.split == "holdout"]
+        train = [t for t in tasks if t.split == "train"]
+        val = [t for t in tasks if t.split == "val"]
+        test = [t for t in tasks if t.split == "test"]
         if limit_replay:
-            replay = replay[:limit_replay]
+            train = train[:limit_replay]
         if limit_holdout:
-            holdout = holdout[:limit_holdout]
-        tasks = replay + holdout
+            val = val[:limit_holdout]
+            test = test[:limit_holdout]
+        tasks = train + val + test
 
     baseline_target = _holdout_hard(target, tasks, skill)