Collect hard disk information

[DeerMaximum]

This pr adds the ability to collect hard disk information.

Lables:

• serial
• health
• devpath
• size
• type
• node

Example:

# HELP pve_disk_info Disk info
# TYPE pve_disk_info gauge
pve_disk_info{devpath="/dev/nvme0n1",health="PASSED",node="hl-cl2",serial="S64*****",size="500107862016",type="nvme"} 1.0
pve_disk_info{devpath="/dev/sda",health="PASSED",node="hl-cl2",serial="S6P*****",size="500107862016",type="ssd"} 1.0

[znerol]

Thanks for taking the time to file a PR.

Unfortunately scraping configuration has turned out to be inherently inefficient (see #55 and #58). Especially for big and growing deployments, this can get nasty quite quickly. In fact I intend to disable the config collector by default in an upcoming release (i.e., --no-collector.config).

This PR is using the exact same known-to-be-faulty mechanism to collect the desired data. To make matters worse, the problematic loop would be running twice after that PR landed and as a result the time to collect all metrics will double for many users.

See the comments in #115 for alternative ideas on how to scrape config efficiently.

[znerol]

Closed, see #156