UME schedule: interval from any finished job; drop trailing sleep; stale running cleanup; paging logs

- Use last ended_at (done or failed) for debounce; remove duplicate time.sleep after sync\n- On startup mark running+no ended_at jobs failed (crash/interrupt) so interval logic is not fooled\n- Log marker page progress every 25 pages\n- Log when alarm/inventory scheduler threads start

Co-authored-by: Cursor <cursoragent@cursor.com>

netx_api/ume_sync_service.py

@@ -132,6 +132,15 @@ def _collect_marker_pages(
         rows = [x for x in rows if isinstance(x, dict)]
         pages.append(rows)
 
+        if page_no == 1 or page_no % 25 == 0:
+            _sync_log.info(
+                "ume marker page=%s rows=%s is_end=%s marker_len=%s",
+                page_no,
+                len(rows),
+                getattr(diag, "is_end_of_reply", None),
+                len(str(getattr(diag, "marker", "") or "")),
+            )
+
         # Protection against repeated pages causing infinite loops.
         cur_sig = "|".join(sorted(_alarm_key(x) for x in rows))
         if cur_sig and cur_sig == last_page_signature: