Add link to synthetic churn dataset on Hugging Face

[osaatcioglu]

Closes the gap reported in #28 — the synthetic dataset used to train the churn models in #11 was never committed to the repo. This adds a link to it on Hugging Face: https://huggingface.co/datasets/Omer1234567/telco-churn-synthetic

Also notes that the generator script isn't seeded, so re-running it won't reproduce the exact same data.

[Copilot]

Pull request overview

This PR updates the churn experiment documentation to address the missing synthetic training dataset by linking to the published dataset on Hugging Face and clarifying that the local generator script is non-deterministic (unseeded), so regenerated data will not exactly match the published dataset.

Changes:

• Add a “Dataset” section with a link to the synthetic churn dataset hosted on Hugging Face.
• Note that re-running the generator script produces different values because it is not seeded.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.