feat: oci genai multi-model, serving mode

[anders-swanson]

📑 Description

Add support for OCI Llama APIs, and OCI dedicated AI clusters.

✅ Checks

• My pull request adheres to the code style of this project
• My code requires changes to the documentation
• I have updated the documentation as required
• All the tests have passed

ℹ Additional Information

[AlexsJones]

I am going to provisionally approve the code, but can you please demonstrate the this working as I am unable to test? Also could you explain the changes, and why they are necessary?

[anders-swanson]

Hi @AlexsJones, thanks for taking a look at the code, and apologies for any delay in my response.

This PR adds the following enhancements to the OCI provider:

• Support the meta vendor of Llama family base LLM models, as previously only the cohere vendor family was supported.
• Support OCI dedicated ai clusters, instead of just base models. This lets users wire up their fine-tuned OCI models with K8sgpt 👍
• Add the new topk parameter to the OCI provider.

Otherwise, the OCI provider works much the same as it did before:

% k8sgpt auth add -b oci -m <my model> -k <my compartment>

oci added to the AI backend provider list

% k8sgpt analyze --explain -b oci
... response from k8sgpt and oci model ...

[anders-swanson]

@AlexsJones do you need additional information to finish your review?

[anders-swanson]

@AlexsJones checking in here, is this PR still waiting for a review or is there something needed on my end?

[anders-swanson]

closing in favor of fixing #1334

[anders-swanson]

see #1337