fix(state): quarantine SessionDB handle after structural corruption

A bare SQLITE_CORRUPT/NOTADB on a live write (not FTS-scoped, not a
replaced file) now sets a sticky per-instance flag: later writes fail
fast with StateDbCorruptError, the handle never reopens after close(),
and close() skips its explicit PASSIVE WAL checkpoint. Gateway and agent
flush paths divert pending transcripts to JSONL/spool like the replaced
case instead of retrying forever.

Field evidence: a handle that kept writing for ~50 minutes after the
first structural error checkpointed 15 pages under the wrong page
numbers on shutdown (page 1 <- messages_fts_trigram_data leaf), turning
"malformed" into "file is not a database".

Refs #90837, #90950, #97940, #89332, #45383

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CNX8rNYHqA5pT4tAGSzXtb

tests/test_state_db_notadb_fail_closed.py

@@ -14,7 +14,7 @@ from unittest.mock import MagicMock
 
 import pytest
 
-from hermes_state import SessionDB, _on_disk_journal_mode
+from hermes_state import SessionDB, StateDbCorruptError, _on_disk_journal_mode
 
 
 class _NotADbOnce:
@@ -42,9 +42,12 @@ class TestFailClosedAfterNotADb:
             reopen = MagicMock()
             monkeypatch.setattr("hermes_state._connect_tracked_db", reopen)
             db._conn = _NotADbOnce(real_conn)
-            with pytest.raises(sqlite3.DatabaseError, match="not a database"):
+            with pytest.raises(sqlite3.DatabaseError, match="not a database") as excinfo:
                 db.create_session(session_id="s2", source="cli", model="test")
             reopen.assert_not_called()
+            # NOTADB on a live write is structural: the handle is quarantined.
+            assert isinstance(excinfo.value, StateDbCorruptError)
+            assert db._db_corrupt is True
         finally:
             db._conn = real_conn
             db.close()