Add new Run Now button + fix on callback

[chaadow]

There are times where we don't want to wait for the event to happen and manually execute/run the event.

This PR introduces:

• a new Run now button next to the Enable/Disable button

  • After carefully reading clockwork source code, using event.run(Time.now) handles the setting of event.last
  • it also calls set_last_run to update redis as well.

• Setting Last run in after_run callback

  • I usually have every(1.minutes, 'job_name') { SomeSidekiqJob.perform_async }
  • What can happen here, is that the block can raise an exception. For my case, it can be redis ( the one for Sidekiq not clockwork_web) is down or Out Of Memory for example. But it can also be any ruby error
  • So setting "last_run" on Clockwork_web's redis after being run is a safer and more accurate way to set it than setting it before it's run, which can be misleading

  Here is a screenshot:

[ankane]

Hi @chaadow, thanks for the PR. However, I'd prefer to keep the functionality minimal and (personally) think it's more useful to see when the job was last started by the scheduler.

[chaadow]

@ankane Imagine these 2 use cases :

• First use case is : you disable the job temporarily ( feature already present in clockwork_web) then you re-enable it, but you want to run the re-enabled job without waiting for say, another 24 hour or so. so the Run now can help in that case.
• Second use case: The job was called, and it delegated to say ActiveJob/Sidekiq. but the sidekiq job itself failed, and you got a Sentry error. You deploy a fix. You know that the job was not run even though clockwork_web says it did ==> so the Run now in this case can help as well.

From an implementation standpoint, it is pretty simple as well, as it simply calls the stored lambda again.

And on a personal level, I've been using my fork in producition, and the Run now has helped us tremendously. a recent use case, is we were rate-limited by an API. so we had to re-run a job to make sure it was properly handled.