Feature for Async API calls

[fewknow]

Currently the only way to have async calls is to use a null_resource and a script. This works, but it doesn't track the state of the async call for purposes of another run in terraform plan and apply.

terracurl fixes this issue for sync calls but not async. If we could add an async poll api endpoint and expected return code then the provider could poll the endpoint until a 200 was returned. This would allow for async calls throughout a system to be made.

If nothing changed in the call on another plan and apply it would simply be skipped, but if the payload changed because it is an update to the resource, then terraform would run the async call again.

I could use this today in a system I am building for a client. I may try to fork and do it on my own, but I am not familiar with writing providers.

Thanks