Prow: Set proper user-agent in config

[alvaroaleman]

The default user-agentis $0/$version, which is pretty useless because it
always ends up being app.binary/$version. This PR changes that instead
be the components name which is useful for:

• Audit log: We can now both use the user-agent or SA to identify
  components
• ManagedFields: Managed fields now have the correct fieldOwner set

This allows for example the use of kubectl-blame:

$ k blame prowjob ebb68c01-792b-11eb-a360-96cd908b0bc6
                                              apiVersion: prow.k8s.io/v1
                                              kind: ProwJob
                                              metadata:
horologium (Update 2021-02-27 13:45:11 -0500)   annotations:
horologium (Update 2021-02-27 13:45:11 -0500)     prow.k8s.io/job: echo-test
                                                creationTimestamp: "2021-02-27T18:45:11Z"
                                                generation: 4
horologium (Update 2021-02-27 13:45:11 -0500)   labels:
horologium (Update 2021-02-27 13:45:11 -0500)     created-by-prow: "true"
app.binary (Update 2021-02-27 13:45:11 -0500)     prow.k8s.io/build-id: "1365734741038338048"
app.binary (Update 2021-02-27 13:45:11 -0500)     prow.k8s.io/id: ebb68c01-792b-11eb-a360-96cd908b0bc6
horologium (Update 2021-02-27 13:45:11 -0500)     prow.k8s.io/job: echo-test
horologium (Update 2021-02-27 13:45:11 -0500)     prow.k8s.io/type: periodic
                                                name: ebb68c01-792b-11eb-a360-96cd908b0bc6
                                                namespace: prow
                                                resourceVersion: "144661945"
                                                uid: 63f44066-411e-4ff3-9ab4-97ddb2ce636d
horologium (Update 2021-02-27 13:45:11 -0500) spec:
horologium (Update 2021-02-27 13:45:11 -0500)   agent: kubernetes
horologium (Update 2021-02-27 13:45:11 -0500)   cluster: default
horologium (Update 2021-02-27 13:45:11 -0500)   decoration_config:
horologium (Update 2021-02-27 13:45:11 -0500)     gcs_configuration:
horologium (Update 2021-02-27 13:45:11 -0500)       bucket: s3://prow-logs
horologium (Update 2021-02-27 13:45:11 -0500)       path_strategy: explicit
horologium (Update 2021-02-27 13:45:11 -0500)     s3_credentials_secret: s3-credentials
horologium (Update 2021-02-27 13:45:11 -0500)     utility_images:
horologium (Update 2021-02-27 13:45:11 -0500)       clonerefs: gcr.io/k8s-prow/clonerefs:v20200608-16190316cf
horologium (Update 2021-02-27 13:45:11 -0500)       entrypoint: gcr.io/k8s-prow/entrypoint:v20200608-16190316cf
horologium (Update 2021-02-27 13:45:11 -0500)       initupload: gcr.io/k8s-prow/initupload:v20200608-16190316cf
horologium (Update 2021-02-27 13:45:11 -0500)       sidecar: gcr.io/k8s-prow/sidecar:v20200608-16190316cf
horologium (Update 2021-02-27 13:45:11 -0500)   job: echo-test
horologium (Update 2021-02-27 13:45:11 -0500)   namespace: test-pods
horologium (Update 2021-02-27 13:45:11 -0500)   pod_spec:
horologium (Update 2021-02-27 13:45:11 -0500)     containers:
horologium (Update 2021-02-27 13:45:11 -0500)     - command:
horologium (Update 2021-02-27 13:45:11 -0500)       - /bin/sleep
horologium (Update 2021-02-27 13:45:11 -0500)       - 1m
horologium (Update 2021-02-27 13:45:11 -0500)       image: alpine
horologium (Update 2021-02-27 13:45:11 -0500)       name: ""
horologium (Update 2021-02-27 13:45:11 -0500)       resources: {}
horologium (Update 2021-02-27 13:45:11 -0500)   report: true
horologium (Update 2021-02-27 13:45:11 -0500)   type: periodic
horologium (Update 2021-02-27 13:45:11 -0500) status:
app.binary (Update 2021-02-27 13:45:11 -0500)   build_id: "1365734741038338048"
app.binary (Update 2021-02-27 13:45:11 -0500)   description: Job triggered.
app.binary (Update 2021-02-27 13:45:11 -0500)   pendingTime: "2021-02-27T18:45:11Z"
app.binary (Update 2021-02-27 13:45:11 -0500)   pod_name: ebb68c01-792b-11eb-a360-96cd908b0bc6
app.binary (Update 2021-02-27 13:45:11 -0500)   prev_report_states:
app.binary (Update 2021-02-27 13:45:11 -0500)     gcsk8sreporter: pending
app.binary (Update 2021-02-27 13:45:11 -0500)     gcsreporter: pending
horologium (Update 2021-02-27 13:45:11 -0500)   startTime: "2021-02-27T18:45:11Z"
app.binary (Update 2021-02-27 13:45:11 -0500)   state: pending
app.binary (Update 2021-02-27 13:45:11 -0500)   url: https://prow.tld/view/s3/s3/prow-logs/logs/echo-test/1365734741038338048

Note: Due to a bug in controller-runtime, this doesn't work for a client
that comes from a Manager that has leader election enabled, like planks
prowjob client above: kubernetes-sigs/controller-runtime#1401
I will bump the dependency ASAP once the upstream fix is in.

[cjwagner]

Neat, TIL about kubectl blame.

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: alvaroaleman, cjwagner

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• prow/OWNERS [alvaroaleman,cjwagner]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment