feat: Llama2 inf2 Ray inference upgrade and bug fix

[vara-bonthu]

What does this PR do?

🛑 Please open an issue first to discuss any significant work and flesh out details/direction - we would hate for your time to be wasted.
Consult the CONTRIBUTING guide for submitting pull-requests.

This PR includes the following updates and fixes for the Llama2 inference setup using Ray on Inf2 instances:

Upgraded Ray Inference Setup:

• Added LIB path to the Docker image path for Ray.
• Updated the Ray Serve YAML configuration.
• Fixed the environment variable setup for neuron cores in the serving script.
• Revised the documentation with accurate deployment steps.
• Added a Gradio Web UI app for Llama2 deployment, making it generic for use with any models by creating the model client as a config map. This approach eliminates the need to create a dedicated image for each model inference, avoiding local Docker image builds.

Motivation

More

• Yes, I have tested the PR using my local account setup (Provide any test evidence report under Additional Notes)
• Mandatory for new blueprints. Yes, I have added a example to support my blueprint PR
• Mandatory for new blueprints. Yes, I have updated the website/docs or website/blog section for this feature
• Yes, I ran pre-commit run -a with this PR. Link for installing pre-commit locally

For Moderators

• E2E Test successfully complete before merge?

Additional Notes

[askulkarni2]

LGTM! Can you please add .py and .yaml files to the ignore list for spell check in the .pre-commit.yaml?

[vara-bonthu]

LGTM! Can you please add .py and .yaml files to the ignore list for spell check in the .pre-commit.yaml?

I am thinking about a way to exclude only the code and still validate the comments under.pyand .yaml files.