Add autoscaling decision to cluster logs

[kunisen]

Issue Description

Currently, when there's an autoscaling on Elastic Cloud Hosted,

The autoscaling capacity API will be called to make the autoscaling decision.
https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-autoscaling-get-autoscaling-capacity

GET /_autoscaling/capacity

However, this log is not able to be seen from cluster logs, and we only see the logs like this:

[instance-00000000XX] adding autoscaling policy [hot_content]

Without having autoscaling capacity information to be recorded in cluster logs, it is hard to understand why the autoscaling was triggered.

Request

It would be great if we could log this info into cluster logs when the endpoint was called.
Then it would be very straightforward and clear to understand why autoscaling was triggered, and can greatly improve user self-service ability.

---

cc @maggieghamry @jakommo

[elasticsearchmachine]

Pinging @elastic/es-distributed-coordination (Team:Distributed Coordination)

[kunisen]

Thanks @dakrone for assigning the team for us.
Do we know what's the level of effort of having this one?

I think it needs some addendum to loggers but we probably don't need some huge code change?
We could provide more case / internal escalation ticket data if these can help with prioritization.
cc @Gunnerva

[kunisen]

Could it be as simple as we add a logger.info to record the reason in below code piece?
https://github.com/elastic/elasticsearch/blob/v9.0.3/x-pack/plugin/autoscaling/src/main/java/org/elasticsearch/xpack/autoscaling/capacity/AutoscalingDeciderResult.java#L61-L73

---

We have autoscaling policy related logger in place but we don't have the one for real action.
https://github.com/elastic/elasticsearch/blob/v9.0.3/x-pack/plugin/autoscaling/src/main/java/org/elasticsearch/xpack/autoscaling/action/TransportPutAutoscalingPolicyAction.java#L166-L167