Add getting up and running docs

Scouter currently lacks clear getting started documentation, making it difficult for new developers to get up and running quickly.

## Tasks

- [x] Write a "Getting Started" guide covering installation, basic setup, and first use
- [x] Document how to spin up the local server (Docker backends + `make start.server`)
- [x] Add a quickstart example for each core feature: drift detection (PSI/SPC/custom), tracing, and GenAI evaluation
- [x] Document environment variables required for the server
- [x] Add a section on the Python client (`ScouterClient`, `Drifter`, `ScouterQueue`)
- [x] Ensure docs are linked from the README

## Notes

- Target audience: developers new to Scouter who want to integrate it into an ML workflow
- Should cover both the Python client and the self-hosted server setup