chore: [DevOps] Fix e2e tests

[Jonas-Isr]

Context

We had 3 e2e tests fail multiple times:

• 2 orchestration tests failed because of a 503 response when using a gemini model with prompt templates or response format
• the model availability test failed due to an internal model not being supported

Feature scope:

• switched usage from gemini model to ChatGPT model for the 2 failing test
  • The underlying issue is a known bug the orchestration team is already working on. This switch of models means we don't have to wait for a fix from their side.
• added a list of internalOnlyModels to the models availability test

Definition of Done

• Functionality scope stated & covered
• Tests cover the scope above
• Error handling created / updated & covered by the tests above
• Aligned changes with the JavaScript SDK
• Documentation updated
• Release notes updated