Sample bundle: ML experiment trail (suggested types: Experiment, Model, Submission, Lesson)

## Summary

The three published sample bundles (ga4, stackoverflow, crypto_bitcoin) are all **data catalog** style — they describe external resources (BigQuery tables, Stack Overflow tags, Bitcoin data). I built a fourth style: an **ML experiment trail** (see bundle at `okf/bundles/s6e2/`).

## Suggested new `type` values for ML domain

The SPEC (`§4.1`) explicitly says types are free, but it would help consumers (visualize, future readers) if a few ML-canonical types were documented:

- `Experiment` — a single training run / iteration (`R0`, `v15f`, etc.)
- `Model` — a trained model (e.g., CatBoost BAG L1, OOF 0.9555)
- `Submission` — a file submitted to a Kaggle LB
- `Lesson` — a discovered or validated skill/principle
- `Competition` — a Kaggle competition as a whole (timeline container)
- `Bundle` — entry point for the whole bundle (top-level `index.md`)

## Why a sample bundle helps

The current GA4 bundle shows OKF for static metadata. An experiment-trail bundle would demonstrate:

1. **Cross-link graph shape** — 11 concepts / 23 edges in my mini-bundle, dense and navigable.
2. **Type-driven color coding in visualize** — already works; would just need ML types documented.
3. **Reference to non-file resources** — `resource: https://github.com/.../pull/5` (PR/issue links) and `resource: file://...` patterns.
4. **Time-evolving knowledge** — `timestamp` field shines here: each run is dated, the dashboard.md can be regenerated from latest `timestamp`.

## My S6E2 mini-bundle (12 concepts, 6 types)

The mini-bundle I built covers an AutoGluon rerun of Kaggle Playground Series S6E2 (Heart Disease). It demonstrates:

- 1 `Bundle` (root)
- 3 `Index` (subdirectory hubs)
- 3 `Experiment` (R0 baseline, R0a bug, R0b fix)
- 3 `Model` (CatBoost BAG L1, LightGBM BAG L1, Weighted Ensemble)
- 2 `Submission` (0/1 thresholded, probability)
- 4 `Lesson` (autogluon-first validated, catboost-first validated, cv-lb-gap refined, submission-format-new)

## Validation

S6E2 (Playground Series S6E2, ended 2026-02-28) — private LB 0.95510 vs user's prior best 0.95516, statistically tied (Δ = -0.00006).

This run also surfaced a new lesson (submission format) that was shipped to `cultivating-ml-agent` as PR #5.

## Checklist

- [x] Bundle follows OKF v0.1 spec (frontmatter + body + file-relative links)
- [x] Includes real failure case (0/1 vs proba, 0.07 LB gap)
- [x] References GitHub PR (cultivating-ml-agent PR #5)
- [x] Uses `resource:` field for both `file://` and `https://`
- [x] Includes 6 different `type` values (Bundle, Index, Experiment, Model, Submission, Lesson)

## Bundle location (in this branch)

`okf/bundles/s6e2/` — run `enrichment-agent visualize --bundle okf/bundles/s6e2` to generate `viz.html`.