Dataloss on daemonset upgrade

## Issue ##

Tested scenario:
- fetch filebeat standalone `curl -L -O https://raw.githubusercontent.com/elastic/beats/8.7/deploy/kubernetes/filebeat-kubernetes.yaml`
- change version to 7.17.6 and set ES connection settings
- `kubectl create -f filebeat-kubernetes.yaml`
- run interactive busybox with a script
- upgrade filebeat config so it points to 7.17.7 and run `kubectl apply -f filebeat-kubernetes.yaml`
- check ES and see logs missing

script used
```
#!/bin/sh
i=0

until [ $i -gt 30000 ]
do
printf "i: %s\n" "$i"
((i=i+1))
i=$((i+1))
sleep 2
done
```

also reproducible on 8.6.2->8.7.0

## Definition of done ##

- [ ] We no more have 0 Filebeat pod at any moment
- [ ] The deamonset upgrade is tested on a K8S infrastructure to validate that logs ingested during the upgrade are all ingested.