Fix pod controller wait logic

[pablochacin]

Description

After changes introduced in #355 the logic for waiting all pod visitors to finish and report errors was broken.

Using a WaitGroup makes the pod controller wait for all pod visitors to end before checking for errors. Consequently, If one pod visitor fails (for instance, the agent command fails) the rest of the visitors will not be immediately canceled.

Checklist:

• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added tests that prove my fix is effective or that my feature works.
• I have run linter locally (make lint) and all checks pass.
• I have run tests locally (make test) and all tests pass.
• I have run relevant integration test locally (make integration-xxx for affected packages)
• I have run relevant e2e test locally (make e2e-xxx for disruptors, or cluster related changes)
• Any dependent changes have been merged and published in downstream modules
  

[roobre]

I think a regular, c-style for loop migth be slightly more readable. WDYT?

Suggested change

	for {
	select {
	case e := <-doneCh:
	if e != nil {
	return e
	}
	pending--
	if pending == 0 {
	return nil
	}
	case <-ctx.Done():
	return ctx.Err()
	}
	}
	for i := 0; i < len(c.targets); i++ {
	select {
	case e := <-doneCh:
	if e != nil {
	return e
	}
	case <-ctx.Done():
	return ctx.Err()
	}
	}
	return nil

[pablochacin]

I'm not convinced. I think that even when this form is simpler, it somehow obscures when the pending is decreased because the select also has the ctx.Done() case.