fix(monitoring): drop deprecated artc0 from cast04 scrape targets

artc0 VM is off; remove node_exporter target and relabel rules from
Prometheus. Point Grafana auth-proxy whitelist and alert-notifier
shortener API at cast01–03 cluster instead of 10.7.16.128.

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

monitoring/cast04/prometheus.yml

@@ -27,7 +27,6 @@ scrape_configs:
           - '10.7.16.237:9100'   # cast02
           - '10.7.16.238:9100'   # cast03
           - '10.7.16.239:9100'   # cast04 (self)
-          - '10.7.16.128:9100'   # artc0 (legacy, monitoring only)
         labels:
           env: 'production'
     relabel_configs:
@@ -47,10 +46,6 @@ scrape_configs:
         regex: '(10\.7\.16\.239):.*'
         target_label: instance
         replacement: 'cast04'
-      - source_labels: [__address__]
-        regex: '(10\.7\.16\.128):.*'
-        target_label: instance
-        replacement: 'artc0'
 
   # ── blackbox_exporter — internal HTTP probes (no TLS issues) ────────────────
   - job_name: 'blackbox_http'