[FEA]: Update examples to use optimized config for `nim_rag_eval_llm`

[yczhang-nv]

Is this a new feature, an improvement, or a change to existing functionality?

Improvement

How would you describe the priority of this feature request

Low (would be nice)

Please provide a clear description of problem this feature solves

Should update AIQ examples to use optimized config for nim_rag_eval_llm:

  nim_rag_eval_llm:
    _type: nim
    model_name: meta/llama-3.1-70b-instruct
    temperature: 0.0000001
    top_p: 0.0001
    max_tokens: 6

Ragas should work better with the config above.

Describe your ideal solution

• Update the example configs as shown above
• Update the related docs

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct
• I have searched the open feature requests and have found no duplicates for this feature request