Add example endless workers

[mariokostelac]

Very often I need some worker to perform basic E2E test. This two workers should represent a worker that's doing IO (sleep) and workers that are CPU intensive (burning metal).
This should help us to:

1. be confident that at least basic e2e flow works when we release a new version
2. write integration tests if we want them (I'd say we do :))
3. do harness testing by providing different environment variables and running shoryuken for long time.
4. help iterating on issues like Log info message when processor dies gracefully #287 quite fast

[phstc]

write integration tests if we want them (I'd say we do :))

We have one. You need SPEC_ALL=true rspec to run it.

How do you test these workers, you start them, then rake console and enqueue messages?

[mariokostelac]

Yes, they're self-serving (creating messages for themselves). I mostly use them to be able to run shoryuken as a long-running process, test that manager dispatching works properly (or at least does not crash) etc. I do not feel strong about having them in the repo, but would simplify manual long-running testing.

[phstc]

@mariokostelac got it! I don't mind having them in the repo if it's part of your testing workflow. But as they are under examples/ would mind to add a comment on top of them explaining it's for testing purposes? Just to avoid misleading people, otherwise they may think it's an example of implementing workers, WDYT? We could also have a specific folder for that, but I don't have strong opinions on that, up to you.

[mariokostelac]

Maybe just putting them under test_workers/?

[phstc]

@mariokostelac sounds good!