Add CNAI Patterns/Blueprints

[zanetworker]

Add a section for patterns/blueprints under https://tag-runtime.cncf.io/wgs/cnaiwg/

Blueprints/patterns are:

• Actionable/Hands-on
• Encapsulates well-accommodated patterns of Cloud Native for (or AI for cloud native).

Patterns/blueprints are are not a replacement to white papers. White papers cover more ground on a large topic (e.g., scheduling, sustainability,...). Blueprints cover a pattern towards a particular outomce. For example, how to scale functions for quick and cost-efficient LLM inferencing.

[pierDipi]

In case this helps with coordinating efforts with other groups, there are 2 proposals related to CNAI blueprints that were discussed in the Kubernetes WG Serving:

1. “Blueprint” Is All You Need: https://docs.google.com/document/d/1tNWnDAhLj9lsTg5dOIBbe7olPH0-4efe2uN-FZIc_KU/edit
2. K8s LLM Serving Catalog: https://docs.google.com/document/d/1yXOhEXltc69_WFX90kMrHs3NcJGsD-U6-RejvK3XQtE/edit

[andreyvelich]

Additionally, here is the KEP from Kubeflow Training WG and Kubernetes Batch WG to create Training Runtimes with blueprints for LLM Fine-Tuning and Distributed Training on Kubernetes: https://github.com/kubeflow/training-operator/pull/2171

[zanetworker]

Very nice, thank you, @andreyvelich and @pierDipi; I will take a closer look at the pointers you shared. This might end up being just the place to consolidate and point to all other blueprints as well (or create new ones for new topics).