Change LH default settings to fix maintenance mode issues (backport #788)

[mergify]

Problem:

The Harvester maintenance mode is blocked by manually attached volumes and single-replica volumes.

Solution:

Change LH default setting to allow node drain on such scenarios.

Refer:

https://github.com/longhorn/charts/blob/v1.6.x/charts/longhorn/values.yaml#L248

https://github.com/longhorn/charts/blob/ad73dc01239b7eeb25ff510ce8358578433d85a5/charts/longhorn/values.yaml#L250

Related Issue:
harvester/harvester#6264
harvester/harvester#6266

Test plan:

Per issue steps.

After a new installation, get such LH default settings:

default-data-path                                                 /var/lib/harvester/defaultdisk                    true      7m19s
...
detach-manually-attached-volumes-when-cordoned                    true                                              true      7m19s
...
node-drain-policy                                                 allow-if-replica-is-stopped                       true      7m19s

Following PRs:

• Upgrade Upgrade: change LH default settings to fix maintenance mode issues harvester#6276
• Document Add troubleshooting of node maintenance mode docs#619

---

This is an automatic backport of pull request #788 done by [Mergify](https://mergify.com).