Add a why chatlas article

[cpsievert]

Based on examples/arguments from https://github.com/jcheng5/llm-quickstart

Could also add a quick example of how much easier it is to get streaming output at the console compared to Pydantic AI

Maybe have this be a dedicated article that we callout and link to from both the README intro and Shiny gen AI docs.

[cpsievert]

There is now a 'why chatlas' section on the homepage.