Assisted inject for datanode migration context

[todvora]

The existing MigrationActionsImpl was a singleton. We were setting the MigrationStateMachineContext to a singleton from different request threads, losing anything that the other thread could have written to the context before. This occured during parallel /state and /trigger request during migrations.

New implementation is using a factory for the MigrationActions, creating new instance for each request, propagating a new instance of MigrationStateMachineContext for each thread. This separates the actions contexts of several threads, never mixing them.

/nocl

Motivation and Context

Sometimes during the connection check in remote reindex migration, a message "No indexes available" has been displayed. There was no problem with the connection check itself, but due to the threading issue and context overriding, we lost the check info and never returned it.

How Has This Been Tested?

Manually + existing set of unit and integration tests.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Refactoring (non-breaking change)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.

[moesterheld]

Very nice solution. just some minor cosmetic changes.