client: expose allocated CPU per task

[hobochili]

Related to #4280 and #5526
Modeled after #5492

Using the currently exposed metrics there is no way to calibrate per-task CPU allocation based on recent usage. This would be helpful, for example, to identify tasks which are greatly over-allocated in terms of CPU.

This PR adds client.allocs.<job>.<group>.<alloc>.<task>.cpu.allocated as a gauge in CPU ticks.

[hashicorp-cla]

All committers have signed the CLA.

[hobochili]

Looks like the lint-go stage timed out. I'm guessing it's a transient error as make check passes locally. Can someone with write permissions please re-run the test(s)?

[tgross]

Thanks @hobochili! LGTM.

We'll want to come back through and clean up the backwards compat sections for 0.11.0, but if we merge this in now it'll be in the 0.10.3 release, so don't worry about that.

[github-actions]

I'm going to lock this pull request because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.