How to categorize this PR?

/area auto-scaling
/area monitoring
/kind enhancement

What this PR does / why we need it:
This PR sets a minAllowed CPU value for prometheus-shoot to 150mCores. We have seen way too frequent evictions on the prometheus-shoot instances for very small absolute CPU values (e.g ~20 changes in 24 hours for moving between 70mCores and 130mCores in intermediate steps of ~20mCores and back again).

We want to address this upstream by adjusting how the vpa computes the upper and lower bounds around the target recommendation. In theory, this is the mechanism that should take care of not evicting for very small changes. In practice, however, the lowerBound and upperBound are way too close to the target, sometimes even identical.

So this is a temporary fix to spend more resources in order to avoid too frequent evictions.

Which issue(s) this PR fixes:
Fixes #

Special notes for your reviewer:
The PR also adds 100M for memory to the minAllowed structure, in order to keep the existing behavior. Currently, no values are defined for VPAMinAllowed, therefore it is defaulted to memory: 100M.

Release note:

Set minAllowed CPU to `150m` for prometheus-shoot to avoid frequent evictions