Problem

The current runtimeOptions.plugins.OpenAI.tqeForInstantTranslateFileTranslation is way to specific and only allows users with human revision roles + edit task roles to have TQE score for pre-translated task. We need more dynamic and flexible way so more users are able to get the score and the translation quality estimation.

Solution

Extend the client-overwritable configuration: runtimeOptions.plugins.OpenAI.tqeForInstantTranslateFileTranslation
Replace its boolean value with a single-choice configuration that controls which InstantTranslate users receive TQE scores for file
translations.

Available options:

• off: TQE disabled for InstantTranslate file translations.
• humanRevision: Only users with both instantTranslate and instantTranslateHumanRevisionAllowed.
• editor: Only users with both instantTranslate and editor.

• humanRevision+Editor: Users with instantTranslate, instantTranslateHumanRevisionAllowed, and editor

• all: All users with the instantTranslate role.

The existing language-resource requirement remains unchanged: TQE is executed only when a matching language resource is assigned to the
selected client and marked as the InstantTranslate TQE default.