Demo: using multiple detectors with llama stack

[m-misiura]

To assist with RHAISTRAT-313 and create a playground that uses TrustyAI's external safety provider for llama stack, it is important to produce representative demos on how to set it up

In this demo, a user is guided through the process of

0. deploying detectors (using HF runtime),
1. specifying these detectors within Guardrails Orchestrator
2. configuring Llama Stack distribution and registering the shield that uses the Guardrails Orchestrator endpoint