Google Vertex AI integration

[silvanocerza]

Summary and motivation

This already mostly implemented but I'm creating this issue to track the remaining work.

Checklist

If the request is accepted, ensure the following checklist is complete before closing this issue.

• The code was merged in the main branch
• Docs: Vertex AI integration docs #155
• There is a Github workflow running the tests for the integration nightly and at every PR
• A label named like integration:<your integration name> has been added to this repo
• The labeler.yml file has been updated
• The package has been released on PyPI
• Create integration tile for Google Vertex AI #120
• The integration has been listed in the Inventory section of this repo README
• There is an example available to demonstrate the feature #117
• The feature was announced through social media #116

[TuanaCelik]

@silvanocerza @masci - since we're also creating an integration directly to Gemini, and as far as we understand this integration has been only to access Gemini models via VertexAI/GCP, does it make more sense to name the generator VertexAIGeminiGenerator in this integration instead of GeminiGenerator?

-- question from @deepset-ai/devrel collectively

[TuanaCelik]

@silvanocerza @dfokina we've created an issue for docs and assigned Daria for now, if that's helpful to track. But I think you should be in touch about the naming of the components here as there's a separate similar integration coming for Makersuite which means the GeminiGenerator might change names