fix: node_md_disks state label as failed

[fsschmitt]

This PR is a mirror of the remote: prometheus#1862

Within node_exporter the metric node_md_disks exports the following states:

• Active
• Failed
• Spare

In the meantime, on the jsonnet alerts it was using the label state="fail" instead of state="failed".

This pull request fixes that issue.

---

Additional information

This information can be also confirmed on:

• Node Exporter collector source code

ch <- prometheus.MustNewConstMetric(
  disksDesc,
  prometheus.GaugeValue,
  float64(mdStat.DisksFailed),
  mdStat.Name,
  "failed",
)

• Node Exporter fixture for node_md_disks

# HELP node_md_disks Number of active/failed/spare disks of device.
# TYPE node_md_disks gauge
node_md_disks{device="md0",state="active"} 2
node_md_disks{device="md0",state="failed"} 0
node_md_disks{device="md0",state="spare"} 0