docs: add known issue for I/O stuck on ARM platforms

[WebberHuang1118]

Problem:

harvester/harvester#10710

Solution:

document update for this known issue

Related Issue(s):

harvester/harvester#10710

Test plan:

Additional documentation or context

[github-actions]

Name	Link
🔨 Latest commit	38a41f7
😎 Deploy Preview	https://6a4b889b67c00700851e5351--harvester-preview.netlify.app

[brandboat]

LGTM, thanks!

[mergify]

Tick the box to add this pull request to the merge queue (same as @mergifyio queue).

• Queue this pull request

[jillian-maroket]

The draft is good. My suggested changes are more structural in nature.

[jillian-maroket]

Suggested change

	On ARM platforms, Longhorn V2 disks that use the SPDK NVMe bdev driver may experience stuck I/O operations. For more information, see [harvester/harvester#10710](https://github.com/harvester/harvester/issues/10710).
	On ARM64 platforms, Longhorn V2 disks that use the SPDK NVMe bdev driver can experience stalled I/O operations. For more information, see issue [#10710](https://github.com/harvester/harvester/issues/10710).

[jillian-maroket]

Suggested change

	1. [Remove the related Longhorn V2 disk](../host/host.md#remove-disks).
	1. [Add the Longhorn V2 disk again](../host/host.md#add-additional-disks).
	1. [Remove the affected Longhorn V2 disk](../host/host.md#remove-disks).
	1. [Add the Longhorn V2 disk again](../host/host.md#add-additional-disks).
	1. Verify that the workaround was successful.
	```
	kubectl -n longhorn-system get blockdevices.harvesterhci.io -o yaml
	```
	The matching `BlockDevice` custom resource should have the following values:
	```
	spec:
	provisioner:
	longhorn:
	diskDriver: aio
	engineVersion: LonghornV2
	status:
	provisionPhase: Provisioned
	state: Active
	```

[jillian-maroket]

Suggested change

	Starting with Harvester v1.9.0, the default disk driver for Longhorn V2 disks is `aio`. Re-adding the disk allows Harvester to use the default driver and avoid the I/O issue on ARM platforms. For more information, see [harvester/harvester#11041](https://github.com/harvester/harvester/issues/11041).
	To verify that the workaround was successful, run the following command:
	```
	kubectl -n longhorn-system get blockdevices.harvesterhci.io -o yaml
	```
	The matching `BlockDevice` custom resource should have the following values:
	```
	spec:
	provisioner:
	longhorn:
	diskDriver: aio
	engineVersion: LonghornV2
	status:
	provisionPhase: Provisioned
	state: Active
	```
	:::note
	Starting with Harvester v1.9.0, the default disk driver for Longhorn V2 volumes is changed to `AIO`. Removing and then re-adding the affected disk forces Harvester to use this new default driver and resolves the stalled I/O operations on ARM64 platforms.
	:::