[Bug] Client may not be reconnecting on server shutdown/restart

[cretz]

Describe the bug

There are reports of clients not auto-reconnecting on server shutdown/restart. Confirm that clients for direct use and for worker use are properly auto-reconnecting. This may be solved by #396, but confirm.

[cretz]

I have done the following test:

• Start a server
• Start a worker
• Start another process that creates a single client at the beginning and runs a workflow every second
• Stop the server
  • This will get core warnings on worker like "WARN temporal_client::retry: gRPC call poll_workflow_task_queue retried 6 times" but will not fail the worker
  • This will get core warnings on the client like "WARN temporal_client::retry: gRPC call start_workflow_execution retried 7 time" and after some retries it will raise an exception to the caller (which we trap for this test)
• Start the server
  • The worker automatically recovers
  • The same exact client instance is usable again for starting workflows

I do not believe this is an issue unless I can get more details on how to replicate