fix(issue-34): review - a correction round keeps the earlier creator jobs' sub-jobs in the evidence table

round_roots was replaced by the correction job, so an implementation
job's open or failed sub-job vanished from the next review and the loop
could pass (planner-reviewer-codex-01, P2). The run's creator jobs now
accumulate; each wait still covers only the new job's tree, so the wait
cap is not paid again per round. T9 covers open / error / completed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

.agents/skills/lib_py/loop/runner.py

@@ -98,7 +98,9 @@ class LoopRunner:
         self._rounds: List[Dict[str, Any]] = []
         self._accepted_round: Optional[Dict[str, Any]] = None
         self.created_jobs: List[str] = []
-        # Issue #34: the creator jobs whose sub-job trees this round's reviewers see
+        # Issue #34: every creator job of this run (execution, corrections,
+        # compliance). Kept across rounds: a later creator job does not make an
+        # earlier one's open or failed sub-job go away - that is the reviewers' call.
         self.round_roots: List[str] = []
         self.wait_open_jobs = False
         self.failed_submit_jobs: Set[str] = set()
@@ -1511,7 +1513,7 @@ class LoopRunner:
 
             correct_jid = self._dispatch(self.creator, creator_type, "Worker", correction_prompt)
             if correct_jid:
-                self.round_roots = [correct_jid]
+                self.round_roots.append(correct_jid)
             if not correct_jid or not self.wait_for_job_tree(correct_jid):
                 log_error("Failed to register Creator correction job.")
                 return 1