Kubernetes Metrics CPU Util returning raw numbers

[icrustandi]

Describe the bug
Metrics gatherer seems to be returning raw CPU values instead of 0-100 for CPU utilization on resource-level metrics.

To Reproduce
Steps to reproduce the behavior:

When setting the following config:

kubernetesMetricSpecs:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization

Metrics gatherer is returning the following:

"kubernetesMetrics": [
    {
        "current_replicas":1,
        "spec": {
            "type":"Resource"
            },
            "resource":{
                "pod_metrics_info":{
                    "myapp-anonymized-123":{
                        "Timestamp":"2025-07-31T17:56:38Z",
                        "Window":59937000000,
                        "Value":4
                        }
                    }
                }
    }
]

This value seems like a discrepancy because I see 50% utilization (it's an 8m CPU request pod):

$ kubectl top pod myapp-anonymized-123
NAME                                                     CPU(cores)   MEMORY(bytes)   
myapp-anonymized-123   4m           37Mi  

The kubectl reading seems to always line up with what is returned from metrics gatherer which makes it seem like it's only returning raw values

Expected behavior
A clear and concise description of what you expected to happen.
Metrics gatherer should return 50 (indicating 50% cpu util)

"kubernetesMetrics": [
    {
        "current_replicas":1,
        "spec": {
            "type":"Resource"
            },
            "resource":{
                "pod_metrics_info":{
                    "myapp-anonymized-123":{
                        "Timestamp":"2025-07-31T17:56:38Z",
                        "Window":59937000000,
                        "Value":50
                        }
                    }
                }
    }
]

Kubernetes Details (kubectl version):
Kubernetes version, kubectl version etc.
v1.25.14