Add an alert history page and tie /health staleness to the poll interval

/events lists every alert from the last 24 hours to 90 days, open or
cleared, with how long each lasted, failed checks and its ticket. Filters
(device, type, severity, state, range) live in the URL, so device pages
and tickets link straight to a device's history. A "Recurring" table
lists alerts that fired more than once on the same target. On phones the
rows become cards.

/api/events/history backs it (100 per page). The existing /api/events is
unchanged.

/health and the dashboard's monitor indicator now call the monitor
stalled after five missed polls (at least 10 minutes) instead of a fixed
20, and /health reports the monitor's age.

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

tests/test_app.py

@@ -316,3 +316,63 @@ class TestCutsheetLivePorts:
     def test_admin_only(self, cs_client):
         client, _ = cs_client
         assert client.get('/api/cutsheet/live-ports', headers=USER).status_code == 403
+
+
+class TestAlertHistory:
+    HISTORY = {'events': [], 'total': 0, 'by_type': {}}
+
+    def _db(self):
+        db = gandalf.db
+        db.get_events_history.return_value = dict(self.HISTORY)
+        db.get_flappers.return_value = []
+        db.get_event_types.return_value = ['host_unreachable']
+        return db
+
+    def test_page_and_nav(self, client):
+        html = client.get('/events', headers=AUTH).get_data(as_text=True)
+        assert 'Alert History' in html and 'events.js' in html
+        assert 'href="/events"' in client.get('/', headers=AUTH).get_data(as_text=True)
+
+    def test_defaults_to_seven_days(self, client):
+        db = self._db()
+        data = client.get('/api/events/history', headers=AUTH).get_json()
+        assert db.get_events_history.call_args.args == (168, '', '', '', 'all')
+        assert data['types'] == ['host_unreachable'] and data['page_size'] == 100
+
+    def test_filters_are_passed_through(self, client):
+        db = self._db()
+        client.get('/api/events/history?range=90d&type=interface_down&target=+large1+&severity=critical'
+                   '&status=resolved&offset=200', headers=AUTH)
+        assert db.get_events_history.call_args.args == (2160, 'interface_down', 'large1', 'critical', 'resolved')
+        assert db.get_events_history.call_args.kwargs == {'limit': 100, 'offset': 200}
+        assert db.get_flappers.call_args.args == (2160, 'interface_down', 'large1', 'critical')
+
+    @pytest.mark.parametrize('query', ['range=1y', 'severity=bad', 'status=open', 'offset=x'])
+    def test_bad_filters_are_rejected(self, client, query):
+        self._db()
+        assert client.get(f'/api/events/history?{query}', headers=AUTH).status_code == 400
+
+    def test_needs_login(self, client):
+        assert client.get('/api/events/history').status_code == 401
+
+
+class TestHealth:
+    def _health(self, client, age_s, poll=120):
+        from datetime import datetime, timedelta, timezone
+        stamp = (datetime.now(timezone.utc) - timedelta(seconds=age_s)).strftime('%Y-%m-%d %H:%M:%S UTC')
+        gandalf.db.get_state.side_effect = lambda key, default=None: stamp if key == 'last_check' else default
+        with patch.object(gandalf, '_cfg', dict(CONFIG, monitor={'poll_interval': poll})):
+            return client.get('/health')
+
+    def test_recent_check_is_healthy(self, client):
+        resp = self._health(client, 150)
+        assert resp.status_code == 200 and resp.get_json()['checks']['monitor'] == 'ok'
+
+    def test_stalled_monitor_fails_after_ten_minutes(self, client):
+        resp = self._health(client, 660)
+        assert resp.status_code == 503
+        assert resp.get_json()['checks'] == {'db': 'ok', 'monitor': 'stale', 'monitor_age_s': 660}
+
+    def test_limit_follows_a_long_poll_interval(self, client):
+        assert self._health(client, 1000, poll=300).status_code == 200
+        assert self._health(client, 1600, poll=300).status_code == 503