Connection's reconnect is asynchronous

[bchamagne]

Description

Reconnect workflow used to be synchronous, which meant all incoming messages were queued to be processed later, and the callers had to wait until eventually they receive a :timeout.

This PR changed the reconnect workflow to be asynchronous, so the callers can get a :closed error message without delay.

NOTE: the first connection is still asynchronous to be able to keep existing code.

Fixes #710

Type of change

• Bug fix

How Has This Been Tested?

Case added to the unit tests

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules