API: Review OTEPS 5 (Global initialization)

[carlosalberto]

The proposed RFC for Global Initialization landed some time ago, and some languages may benefit more than others from it. Nevertheless, wanted to drop a note here on the need to re-review it and make sure we are properly aligned.

For one, we do support a way to manually/explicitly register the global components, but this is also a call to potentially remove the automatic registration (in case it becomes a burden or similar situation).

https://github.com/open-telemetry/oteps/blob/master/text/0005-global-init.md

[carlosalberto]

@Oberon00 perhaps you can take a look at this, as you are the original author? ;)