Fix nodepool creation

[scaliby]

Fixes

When cluster is re-created but nodepool numbers are not consecutive numbers and desired node pool size is smaller than the original one, XPK removes nodepools that are above the specified limit.

For instance, when user creates a cluster with --num-slices=3, it creates a following node pools:

+ 'cluster-np-0',
+ 'cluster-np-1',
+ 'cluster-np-2',

Then, a user manually deletes a node pool on the UI or through gcloud command so the list of node pools is as follows:

'cluster-np-0',
- 'cluster-np-1',
'cluster-np-2',

Then, as a user re-runs cluster create with --num-slices=2, XPK will do propose following changes:

'cluster-np-0',
+ 'cluster-np-1'
- 'cluster-np-2'

Where no changes are expected.

After this change XPK will propose no changes to node pools in the mentioned case.

Testing / Documentation

Change has been covered with unit tests.

• [y] Tests pass
• [y] Appropriate changes to documentation are included in the PR