tiup-clean --log add warining

[glkappe]

Feature Request

Is your feature request related to a problem? Please describe:

The current clean --log implementation will close the instance and clean up *log, which can also be dangerous in the online environment.
We need some tips to tell users about these risks

[tidb@node4107 ~]$ tiup cluster clean qh --log Starting componentcluster`: /home/tidb/.tiup/components/cluster/v1.3.0/tiup-cluster clean qh --log
This operation will clean tidb v4.0.9 cluster qh's log.
Nodes will be ignored: []
Roles will be ignored: []
Files to be deleted are:
172.16.4.107:
/home/tidb/qh_new/deploy/grafana-43000/log/.log
/home/tidb/qh_new/deploy/prometheus-49090/log/.log
/home/tidb/qh_new/deploy/tiflash-19002/log/.log
/home/tidb/qh_new/deploy/tidb-24000/log/.log
/home/tidb/qh_new/deploy/tikv-40164/log/.log
/home/tidb/qh_new/deploy/pd-12379/log/.log
172.16.4.237:
/home/tidb/qh_new/deploy/tidb-24000/log/*.log
Do you want to continue? [y/N]: y
Cleanup cluster...

• [ Serial ] - SSHKeySet: privateKey=/home/tidb/.tiup/storage/cluster/clusters/qh/ssh/id_rsa, publicKey=/home/tidb/.tiup/storage/cluster/clusters/qh/ssh/id_rsa.pub
• [Parallel] - UserSSH: user=tidb, host=172.16.4.107
• [Parallel] - UserSSH: user=tidb, host=172.16.4.237
• [Parallel] - UserSSH: user=tidb, host=172.16.4.107
• [Parallel] - UserSSH: user=tidb, host=172.16.4.107
• [Parallel] - UserSSH: user=tidb, host=172.16.4.107
• [Parallel] - UserSSH: user=tidb, host=172.16.4.107
• [Parallel] - UserSSH: user=tidb, host=172.16.4.107
• [ Serial ] - StopCluster
  Stopping component grafana
  Stopping instance 172.16.4.107
  Stop grafana 172.16.4.107:43000 success
  Stopping component prometheus
  Stopping instance 172.16.4.107
  Stop prometheus 172.16.4.107:49090 success
  Stopping component tiflash
  Stopping instance 172.16.4.107
  Stop tiflash 172.16.4.107:19002 success
  Stopping component tidb
  Stopping instance 172.16.4.237
  Stopping instance 172.16.4.107
  Stop tidb 172.16.4.237:24000 success
  Stop tidb 172.16.4.107:24000 success
  Stopping component node_exporter
  Stopping component blackbox_exporter
  Stopping component tikv
  Stopping instance 172.16.4.107
  Stop tikv 172.16.4.107:40164 success
  Stopping component pd
  Stopping instance 172.16.4.107
  Stop pd 172.16.4.107:12379 success
  Stopping component node_exporter
  Stopping component blackbox_exporter
• [ Serial ] - CleanupCluster
  Cleanup instance 172.16.4.107
  Cleanup 172.16.4.107 success
  Cleanup instance 172.16.4.237
  Cleanup 172.16.4.237 success
  Cleanup cluster qh successfully

`
Describe the feature you'd like:

There may be a way to clean up the current and archived logs gracefully (maybe individual components do not have log rotation)

Describe alternatives you've considered:

Teachability, Documentation, Adoption, Migration Strategy:

[lucklove]

Currently, TiUP will stop the cluster before cleanup logs, because the components will be writing the log if we don't stop them. And if the log file is hold by the component and we delete it, the component will still write it after the file removed. And the disk usage increase but no useful log write down.