Make nodepool concurrent ops scale better

[modular-magician]

The purpose of this change is to stagger operation creations, but still keeps operations concurrent, so that operations are not bottlenecked on creation and fail.
To achieve the above,

• a cluster-scope lock is held for operation creation
• cluster-scope lock is released after creation is done
• cluster-scope read lock is held while waiting for operation to complete (as before)

Release Note Template for Downstream PRs (will be copied)

See Write release notes for guidance.

container: make nodepool concurrent operations scale better

Derived from GoogleCloudPlatform/magic-modules#12488

[github-actions]

I'm going to lock this pull request because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.
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.