Examples of end-2-end orchestrator deployments

[m-misiura]

I've created instructions for how to deploy the orchestrator stack with guardrails being Hugging Face AutoModelForSequenceClassification models and the generation model model being a Hugging Face AutoModelForCausalLM model

These instructions can be found in a newly created subdirectory end-to-end inside the guardrails. The folder structure is as follows

There are three model serving runtimes under consideration:

1. caikit-nlp
2. caikit-tgis
3. tgis
4. vllm