Move port config back to viper

[motatoes]

#1464 ships a quick fix for #1456 port issue. However there is an underlying problem with viper port loading that I was not able to rootcause yet which is causing port to be pulled as "0" be default from the config object. This has only been reproduced in a k8s environment with a config such as:

apiVersion: v1
kind: Service
metadata:
  name: digger
  namespace: digger
spec:
  ports:
  - port: 80
    targetPort: 3000
    protocol: TCP
  selector:
    app: digger
---
apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: digger
  name: digger
  namespace: digger
spec:
  replicas: 1
  selector:
    matchLabels:
      app: digger
  template:
    metadata:
      labels:
        app: digger
    spec:
      containers:
      - env:
        - name: HOSTNAME
          value: http://localhost:3000
        - name: well
          value: "abc"

        - name: ALLOW_DIRTY
          value: "true"
        - name: GITHUB_ORG
          value: xxxx
        - name: BEARER_AUTH_TOKEN
          value: sadfasdfasd
        - name: DATABASE_URL
          value: postgres://postgres:root@mypostgresinstance/postgres?sslmode=disable
        - name: HTTP_BASIC_AUTH
          value: "1"
        - name: HTTP_BASIC_AUTH_USERNAME
          value: xxxx
        - name: HTTP_BASIC_AUTH_PASSWORD
          value: "sdfasdfsdf"            
        - name: DIGGER_MAX_CONCURRENCY_PER_BATCH
          value: "1"
        image: registry.digger.dev/diggerhq/digger_backend:v0.4.26
        name: backend
        ports:
        - containerPort: 3000

And when the service starts it is being loaded as port "0" when we try to fetch it from viper config - currently only reproduced in a k8s environment but not in docker-compose or binary environments. Need to dig into the viper config/config.go and figure out whats going on causing this 0 value in port.