DeregisterInstanceTraceEvent should always hold instance port

[zqq90]

Describe the bug

When the heartbeat of an instance times out, it will be deregistered, and a DeregisterInstanceTraceEvent will be published,
However, this event does not hold the port of the instance

Expected behavior

Hold hold instance port, Consistent with 'client initiates request'

See:

nacos/naming/src/main/java/com/alibaba/nacos/naming/controllers/InstanceControllerV2.java

Line 159 in fcc9b0a

instance.toInetAddr()));

Actually behavior

See:

nacos/naming/src/main/java/com/alibaba/nacos/naming/healthcheck/heartbeat/ExpiredInstanceChecker.java

Line 82 in fcc9b0a

false, DeregisterInstanceReason.HEARTBEAT_EXPIRE, service.getNamespace(), service.getGroup(), service.getName(), instance.getIp()));

[scienceZ]

thank you so much for your suggestion，the instanceIp should hold instance port. I will check this problem in all trace module.

[zqq90]

thank you so much for your suggestion，the instanceIp should hold instance port. I will check this problem in all trace module.

The field instanceIp is ambiguous. It is recommended to add a new field, or rename it.

cc @scienceZ

[zqq90]

TO @KomachiSion

The affected functions seem to be newly added in this milestone.
So can this issue catch up with version 2.1.1?

[KomachiSion]

Has Pr #8805 fixed this problem?

[scienceZ]

Has Pr #8805 fixed this problem?

yes