Fixes https://emqx.atlassian.net/browse/EMQX-11086
There’s currently a metric inconsistency due to the internal buffering nature of Kafka
Producer (wolff).
We use simple_sync_query to call the Kafka Producer bridge. If that times out, the call
is accounted as failed, even though the message is buffered in wolff and later sent
successfully.
We need to reverse the dependency of `emqx_bridge` and `emqx_bridge_*`, because the former
loads and starts bridges during its application startup. If the individual bridge
application being loaded has not started with its dependencies, the supervision tree will
not be ready for that.
Using `simple_one_for_one` has a potential race condition issue where
we read the PID of the resource manager before trying to remove a
resource, and then that PID changes because it was either dead at
first, or it crashed and changed, and later we use this stale PID to
try to remove it from the supervisor. Under such circumstances, the
restarting child might linger in the supervisor, leaking resources.
By using the resource ID itself as a child ID (and using `one_for_one`
restart strategy), we ensure the child is truly removed.
Andrew Mayorov