fix(kubevirt): Improve error handling in migration monitor during and after live migration

[yaroslavborbat]

Description

The migration monitor (migrationMonitor) continues to make calls to GetJobStats() even after a successful migration has completed. This leads to errors like:

"Failed to get domain job info": virError(Code=42, Domain=10, Message='Domain not found...')

These errors occur because the domain has already been successfully migrated and cleaned up on the source node.

Additionally, during migration, there are cosmetic errors such as:

"Requested operation is not valid: domain is not running"

This PR resolves the issue by properly handling domain lifecycle events during and after migration.

Why do we need it, and what problem does it solve?

What is the expected result?

Checklist

• [ ] The code is covered by unit tests.
• [ ] e2e tests passed.
• [ ] Documentation updated according to the changes.
• [x] Changes were tested in the Kubernetes cluster manually.

Changelog entries

section: vm 
type: fix
summary: Improve error handling in migration monitor during and after live migration
impact_level: low