runtime: Allocate default workload vcpus

[manuelh-dev]

Merge Checklist

• Followed patch format from upstream recommendation: https://github.com/kata-containers/community/blob/main/CONTRIBUTING.md#patch-format
  • Included a single commit in a given PR - at least unless there are related commits and each makes sense as a change on its own.
• Aware about the PR to be merged using "create a merge commit" rather than "squash and merge" (or similar)
• The upstream/missing label (or upstream/not-needed) has been set on the PR.

Summary

• Let's align assignment with the new mem assignment. Extend our 'memory allocation patch' with 'vcpu'. This way we can set default_vcpus to 0 and add 1 vCPU if no limits are given, similar to pull/281.
• Let's allow for non-integer vCPU limits and round up - no change required. the calculation is already as of today for the example of a limit of 3500millis: ceil(0, 3.5) = 4
• We will not hit the bug where we can ask CLH to allocate 5 vCPUs, then hit a crash. We did only previously hit this through ceil(1+3.5)=5. To prevent this we would have had to define a vCPU limit of 1 or higher (cause then kubelet will check 1+3.5 vs 3.86 and tell use that we can't run the pod. Without this, kubelet thinks we only want 3.5 and let's use go ahead and let's CLH allocate a VM with 5 vCPUs, and then 'boom'). 
	○ With this, we do not need to define a vCPU overhead.

Test Methodology