disk-smart: Divide into smaller checks

[markuslf]

Is your feature request related to a problem? Please describe.
Example: When you receive a SMART error in the disk's "Error Log" (which can happen quite often), the check reports CRIT. If you later receive more SMART errors in any attribute, you will not notice it, because the check is already in CRIT status.

Describe the solution you'd like
Divide the disk-smart check into smaller, more specialized parts:

• disk-smart-general
• disk-smart-attributes
• disk-smart-stats
• etc.