TrustyAI bias and drift labs

[erzhan46]

Trusty AI labs:
bias metrics - from https://github.com/trustyai-explainability/odh-trustyai-demos/tree/main/2-BiasMonitoring/kserve-demo
drift metrics - from https://github.com/trustyai-explainability/odh-trustyai-demos/tree/main/3-DataDrift/kserve-demo

Topics for considerations:

• Licensing/credits for use of demos from https://github.com/trustyai-explainability/odh-trustyai-demos
• Use of custom image containing models for labs (http://quay.io/trustyai_testing/modelmesh-minio-examples:latest)
• Use of custom runtime images:
  http://quay.io/opendatahub/openvino_model_server:stable-nightly-2024-08-04
  http://quay.io/trustyai_testing/mlserver:1.6.1

[erzhan46]

@strangiato @carlmes
When you get a chance - could you review this PR?

[jland-redhat]

So I just had a few questions here.

First is, do we want demo data for trusty inside of this repo? Would it make more sense to pull the actual data from somewhere else?

And secondly do we want to consider adding the trusty data to our helm chart repo and referencing that so we can have it in a reusable state?

I know that would be a little extra work so maybe that is a future plan.

What do you think @strangiato @carlmes ?