incorrect timeout parsing

We are leveraging helm-controller via rke2 in harvester, and have noticed that the timeout duration is not parsed correctly when being passed as env variables to the job

for example, sample HelmChart CRD

apiVersion: helm.cattle.io/v1
kind: HelmChart
metadata:
  creationTimestamp: "2022-09-29T03:32:25Z"
  finalizers:
  - wrangler.cattle.io/helm-controller
  generation: 2
  name: vm-import-controller
  namespace: harvester-system
  ownerReferences:
  - apiVersion: harvesterhci.io/v1beta1
    kind: Addon
    name: vm-import-controller
    uid: 5cfce8e3-70f8-48ae-840e-e1081812a9f2
  resourceVersion: "106365"
  uid: 97e7da21-9def-402a-94f2-f6254802ec06
spec:
  chart: harvester-vm-import-controller
  repo: http://harvester-cluster-repo.cattle-system.svc/charts
  timeout: 10m0s
  valuesContent: |
    image:
      tag: v0.1.0
  version: 0.1.1-rc1
status:
  jobName: helm-install-vm-import-controller

The generated job is as follows:

   spec:
      containers:
      - args:
        - install
        - --repo
        - http://harvester-cluster-repo.cattle-system.svc/charts
        - --version
        - 0.1.1-rc1
        env:
        - name: NAME
          value: vm-import-controller
        - name: VERSION
          value: 0.1.1-rc1
        - name: REPO
          value: http://harvester-cluster-repo.cattle-system.svc/charts
        - name: HELM_DRIVER
          value: secret
        - name: CHART_NAMESPACE
          value: harvester-system
        - name: CHART
          value: harvester-vm-import-controller
        - name: HELM_VERSION
        - name: TARGET_NAMESPACE
          value: harvester-system
        - name: TIMEOUT
          value: '&Duration{Duration:10m0s,}'
        - name: NO_PROXY
          value: .svc,.cluster.local,10.52.0.0/16,10.53.0.0/16
        - name: FAILURE_POLICY
          value: reinstall
        image: rancher/klipper-helm:v0.7.3-build20220613
        imagePullPolicy: IfNotPresent
        name: helm

k3s-io / helm-controller

incorrect timeout parsing #160