PD nodes status are not stable if some nodes are down

[baurine]

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. What did you do?

I deploy a cluster with 3 PD nodes, 1 TiDB node, 1 TiKV node by vagrant, then I pause a PD node.

2. What did you expect to see?

I expect to see PD nodes status with 2 healthy and 1 unhealthy.

3. What did you see instead?

The PD nodes statuses are not stable with different output each time after running tiup cluster display command, while they are normal in the tidb-dashboard and the response of API /pd/health.

4. What version of TiUP are you using (tiup --version)?

v0.0.5 tiup with v1.0.3 tiup cluster

[baurine]

I will submit PR to resolve it.