[Website]: Improve docs for using Logstash output with Fleet (_id situation)

### Before you submit

- [x] This issue is about a documentation page, flow, or piece of content.

### Type of issue

Other

### What documentation page or section is affected

https://www.elastic.co/docs/reference/fleet/logstash-output

### What happened?

When using the Logstash output for Elastic Agent, in the output we recommend:
```
output {

if [@metadata][_id] {
elasticsearch {
...
document_id => "%{[@metadata][_id]}"
...
}
} else {
elasticsearch {
...
}
}
```

It is necessary because SOME but not all the integrations use `_id` to avoid duplication.

The drawback is this config duplicates the ES output and can lead to sub-optimal ingestion (due to bulk sizes might vary a lot in size).

A possible optimization for Logstash 8.12+ / or ES Output plugin 11.22.2 would be to leverage the undocumented `[@metadata][_ingest_document][id]`

The recommended config can become:
```
filter {
if [@metadata][_id] {
mutate { add_field => { "[@metadata][_ingest_document][id]" => "[@metadata][_id]" } }
}
}
output {
elasticsearch {
...
}
}
```

### Additional info

I would like @robbavey to review this.

The field `[@metadata][_ingest_document][id]` is not an actual public contract, but it might be useful to simplify the ES output.