[containerinsights] Update GPU usage metrics emitted

[sky333999]

Description of changes

Updating metric declarations for GPU request, limit, usage & reserved_capacity metrics to more closely resemble CPU metrics. Depends on amazon-contributing/opentelemetry-collector-contrib#225.

• Renaming (pod|node)_gpu_total to (pod|node)_gpu_usage_total
• Adding (pod|node)_gpu_reserved_capacity metrics
• Dropping cluster_gpu_* metrics since those can be derived using the [ClusterName] aggregation on the node_gpu_* metrics.
• node_gpu_request will exist in the EMF entry, but not be extracted as a metric - similar to node_cpu_request.

License

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

Tests

• Deployed changes to a cluster running both GPU and non-GPU nodes and validated the EMF log events look as expected.
• New metrics being extracted on a EMF log of Type Pod for a GPU pod:

{
  "CloudWatchMetrics": [
    {
      "Namespace": "ContainerInsights",
      "Dimensions": [
        [
          "ClusterName",
          "Namespace",
          "PodName"
        ],
        [
          "ClusterName"
        ],
        [
          "ClusterName",
          "FullPodName",
          "Namespace",
          "PodName"
        ]
      ],
      "Metrics": [
        {
          "Name": "pod_gpu_reserved_capacity",
          "Unit": "Percent"
        },
        {
          "Name": "pod_gpu_limit",
          "Unit": "Count"
        },
        {
          "Name": "pod_gpu_usage_total",
          "Unit": "Count"
        },
        {
          "Name": "pod_gpu_request",
          "Unit": "Count"
        }
      ]
    }
  ]
}

• New metrics being extracted on a EMF log of Type Node for a GPU node:

"CloudWatchMetrics": [
    {
      "Namespace": "ContainerInsights",
      "Dimensions": [
        [
          "ClusterName"
        ],
        [
          "ClusterName",
          "InstanceId",
          "NodeName"
        ]
      ],
      "Metrics": [
        {
          "Name": "node_gpu_reserved_capacity",
          "Unit": "Percent"
        },
        {
          "Name": "node_gpu_limit",
          "Unit": "Count"
        },
        {
          "Name": "node_gpu_usage_total",
          "Unit": "Count"
        }
      ]
    }
  ]

Requirements

Before commit the code, please do the following steps.

1. Run make fmt and make fmt-sh
2. Run make lint