[VC-35738] Allow DataGatherer.Run to return without stopping the other errgroup Go Routines

[wallrj]

In #601 (comment) I introduced a bug which caused the API server and VenafiConnection helper to exit as a result of the early return of some DataGatherer.Run implementations.

Tested using hack/e2e/test.sh.

Before this change, the venctl installation part of the script failed,
because the healthz server gets quickly shutdown as a result of the Discovery DataGatherer.Run function returning and cancelling the errgroup.
This causes the Liveness check to fail, and the Helm install never succeeds.

After this change, the test script works again:

$ ./hack/e2e/test.sh
...
{
  "ts": 1730909040560.407,
  "caller": "logs/logs.go:153",
  "msg": "successfully gathered 10 items from \"k8s/namespaces\" datagatherer",
  "source": "agent",
  "v": 0
}
{
  "ts": 1730909040560.5063,
  "caller": "logs/logs.go:153",
  "msg": "successfully gathered 0 items from \"k8s/googlecasclusterissuers\" datagatherer",
  "source": "agent",
  "v": 0
}
{
  "ts": 1730909040560.5884,
  "caller": "logs/logs.go:153",
  "msg": "successfully gathered 0 items from \"k8s/gateways\" datagatherer",
  "source": "agent",
  "v": 0
}
{
  "ts": 1730909040560.6826,
  "caller": "logs/logs.go:153",
  "msg": "Posting data to: ",
  "source": "agent",
  "v": 0
}
{"ts":1730909041447.9111,"caller":"logs/logs.go:153","msg":"Data sent successfully.","source":"agent","v":0}

[maelvls]

Side-note, you don't need to change anything:

I find it weird that the Run func's implementation varies from data gatherers to data gatherers...

The specification of the Run func (an interface, I assume) seems to be defined as "Run blocks until the channel is closed".

But the local datagatherer, for example, is implemented like this:

func (g *DataGatherer) Run(stopCh <-chan struct{}) error {
	// no async functionality, see Fetch
	return nil
}

It should be:

func (g *DataGatherer) Run(stopCh <-chan struct{}) error {
	// no async functionality, see Fetch
	<-stopCh
	return nil
}