collector: support custom poll intervals

Also simplify config context schema for SNMP since we only use a single
instance.

roles/collector/templates/prometheus-job.yml.j2

@@ -1,10 +1,14 @@
-{% set devices = query("netbox.netbox.nb_lookup", "devices", api_filter="{{ item.nb_filter }}", raw_data=true)
+{% set devices = query("netbox.netbox.nb_lookup", "devices", api_filter="{{ item.hosts }}", raw_data=true)
     | selectattr("primary_ip")
     | map(attribute="name")
     | map("extract", hostvars) -%}
 
 scrape_configs:
   - job_name: "{{ item.name }}"
+{% if item.interval is defined %}
+    scrape_interval: {{ item.interval }}s
+    scrape_timeout: {{ item.interval // 5 }}s
+{% endif %}
     relabel_configs:
       - source_labels: [__address__]
         regex: '([^.]+).*'