Add `check` command to check the status of the target host

[zyy17]

Background

For some scenarios, we want to know some basic information about the target host:

• CPU
• Memory
• Disk(maybe can use fio)
• Network

When the check command is finished, we can get the basic report of the target host.

btw: I'm unsure whether we should include the check command in the cluster subsystem or create a separate subsystem named diagnose.

[sjcsjc123]

Can we ssh the target host? Or send a rpc/grpc request to the target host?

[zyy17]

@sjcsjc123 I think it's OK to do that. Actually, I don't have a clear design now, and it's still the idea. From my experience, I think we should consider the scenario of K8s and bare metal.