This PR introduces the following changes:

• fix: Report LB pool health properly

  The cloudflare_zone_pool_health_status metric introduced with goal
  to report the actual pools health status during the time, regardless
  if this pool is the selected one or not. The initial implementation
  uses the loadBalancingRequestsAdaptiveGroups type which returns
  aggregated Load Balancing origin requests with adaptive sampling and
  contains information only about the selectedPool (Name, Health,
  AvgRttMs, etc), rather than for all LB pools. This leads on reporthing
  back the status only of the selectedPool (which normally is always 1)
  and not the overall status of all LB's pools.
  According to Cloudflare's documentation about LoadBalancing GraphQL
  Analytics API[1], the loadBalancingRequestsAdaptive schema can be
  used to fetch analytics about the Raw Load Balancing origin requests
  with adaptive sampling. These analytics exposes infromation about the
  selected pool like:

  • Selected Pool - Name/Health/ID/Average RTT
  • Session Affinity - Type/Status

  But it also exposes information about LB's pools and origins regardless
  the selection decision, like:

  • Pools ID/Name/Health/Average RTT/Selected
  • Origins Name/FQDN/Ipv4/Ipv6/Selected

  The latter metrics can be used to report the health of all pools
  associated with account's LBs.
  For this reason, this commit adds the required loadBalancingRequestsAdaptive
  schema struct on the lbResp struct and configures it as the source for the
  cloudflare_zone_pool_health_status metric.

  1. https://developers.cloudflare.com/load-balancing/reference/load-balancing-analytics/#graphql-analytics

• feat: Add poolRequestsTotal metric

  This commit introduces a prometheus counter metric about the total
  number of requests per pool.