Core: cancel moved runs, retry bookkeeping, progress cap, safer classification

- A run whose item leaves In Progress is killed, gets a [worker] cancel comment,
  frees its terminal slot and serialize group and keeps its thread id; if the
  kill fails the slot stays taken and the cancel is retried next tick.
- Transient retries are the same logical round: they do not count toward
  max_rounds_per_thread and do not trigger a max-rounds rebuild.
- New project setting max_progress_rounds (default 12): after that many
  consecutive [progress] rounds the item goes to Waiting (answerer) with a note.
- retry/rebuild markers are consumed only after the run launched, so a failed
  start no longer leaves an In Progress item that is never picked again; a
  rebuild is no longer reported as a retry.
- State is saved after every finished round and every launched run.
- Failure classification: transient patterns only look at the tail, and an
  exit-0 failure is logic (never retried) unless the tail shows a transient error.
- First-line tag: tolerate BOM and a leading code fence / rule line.
- Start comment notes when a handoff summary was attached.

tests/test_prompt.py

@@ -37,6 +37,13 @@ class PromptTest(unittest.TestCase):
         self.assertIn("RETRY", p)
         self.assertNotIn("没有新评论", p)
 
+    def test_new_comment_filter_is_by_id_not_timestamp(self):
+        late_but_old = [{"id": "c9", "created_at": "t0", "comment_html": "<p>clock-skewed reply</p>"},
+                        {"id": "c1", "created_at": "t9", "comment_html": "<p>edited old one</p>"}]
+        p = build_prompt("GW-1", ISSUE, late_but_old, {"c1"}, True)
+        self.assertIn("clock-skewed reply", p)
+        self.assertNotIn("edited old one", p)
+
     def test_handoff(self):
         rec = {"runs": 3, "tags": ["progress", "progress", "question"], "last_message": "[question] A or B?"}
         h = handoff_summary(rec, "线程已续 3 轮")