Support async processing on hosting where workers are not available

This MR also contains code from ✨ Implement retry Active because it needs support for async transports

We need to postpone this on getting the SQL side in first.

The code right now is a spaghetti mess.

I'm not sure about the premise of this issue. To date, I've recommended using the limit=Xseconds command combined with traditional cron if the platform is intolerant of long running commands. Why is that not a viable option for Platform.sh?

Secondly, if a platform doesnt support async, then perhaps they should just consider using synchronous sending?

Lateruntime/kernel-termination processing of async messages, or "poor-mans-async" is a bit questionable. I'm wondering if that should just be a different contrib, since its not something we should ever be endorsing.