fix(worker): cap slskd ETA + rollback on progress-write failure (stop txn-abort job churn)

A near-stalled Soulseek peer drives the EWMA speed toward zero, so
_eta_seconds returned int(remaining/speed) values far past int4 max. Writing
that to Job.downloadEtaSeconds (an integer column) raised "integer out of
range", which aborted the SHARED pipeline connection's transaction. The
on_progress except clause logged but never rolled back, so every subsequent
query cascaded "current transaction is aborted" and the whole job failed and
requeued (9 such failures / 24h observed in prod).

- Cap _eta_seconds at ~100h (359999s), well inside int4.
- Roll back the shared conn in on_progress's except so a failed progress
  write can never poison the pipeline txn, honoring the existing docstring
  promise that "a progress write must never fail the download".
- Regression test for the near-stall cap.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

worker/lyra_worker/pipeline.py

@@ -191,6 +191,12 @@ def _make_on_progress(conn: psycopg.Connection, job_id: str, reports: "threading
             try:
                 _set_download_progress(conn, job_id, frac, eta_seconds)
             except Exception as e:  # a progress write must never fail the download
+                # Roll back so a failed write can't leave the shared pipeline conn in an aborted
+                # txn (which would cascade "current transaction is aborted" into every later query).
+                try:
+                    conn.rollback()
+                except Exception:
+                    pass
                 print(f"pipeline: on_progress write failed: {e}", flush=True)
 
     return on_progress