docs: add BootstrapODPSample doctest examples

[priyam0k]

Adds a Sphinx doctest Examples section to the BootstrapODPSample class.

Refs #704 (Stochastic / Simulation → BootstrapODPSample row).

Summary of Changes

• Adds an Examples block to BootstrapODPSample in chainladder/adjustments/bootstrap.py.
• Three .. testcode:: / .. testoutput:: groups covering:
  1. Basic fit on the RAA sample — shows resampled_triangles_.shape and scale_.
  2. Downstream stochastic IBNR — Chainladder().fit(sims).ibnr_.sum('origin') aggregated to mean / std across 100 simulations.
  3. Parameter effect — drop_high=True shrinks scale_.
• Uses random_state=42 and n_sims=100 so output is deterministic, fast, and stable across CI runners.
• Docstring-only; no runtime logic or public API changes.

Related GitHub Issue(s)

• Refs API Reference Examples #704
• Follows the conventions used in DOCS: Expand Triangle constructor examples (#704) #714, DOCS: Add examples to Triangle methods and properties (#704) #719, and the in-flight docs: add tail estimator doctest examples #801.

Additional Context for Reviewers

• Doctest CI passed locally and on GitHub Actions (Doctest (ubuntu-latest, 3.12)).

• pytest chainladder/adjustments/tests -q → 16 passed.

• Intentionally excludes .github/workflows/sync-main-to-docs.yml, matching the split-PR convention from docs: add tail estimator doctest examples #801.

• Open to feedback on whether the IBNR example should also surface total_process_risk_ / total_parameter_risk_ from a paired MackChainladder view — happy to extend in a follow-up if useful.

• I passed tests locally for both code (uv run pytest) and documentation changes (uv run jb build docs --builder=custom --custom-builder=doctest)

---

Note

Low Risk
Docstring-only documentation change; the only risk is doctest brittleness if numeric outputs or sampling behavior change across environments.

Overview
Adds a new Examples section to the BootstrapODPSample docstring with Sphinx doctests.

The examples show (1) deterministic fit output (resampled_triangles_.shape and scale_) on the RAA sample, (2) feeding bootstrap samples into Chainladder to summarize simulated IBNR mean/std, and (3) how drop_high=True changes the resulting scale_.

^{Reviewed by Cursor Bugbot for commit 48ab897. Bugbot is set up for automated code reviews on this repo. Configure here.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.94%. Comparing base (e76d85d) to head (48ab897).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #836   +/-   ##
=======================================
  Coverage   86.94%   86.94%           
=======================================
  Files          86       86           
  Lines        4994     4994           
  Branches      644      644           
=======================================
  Hits         4342     4342           
  Misses        462      462           
  Partials      190      190           

Flag	Coverage Δ
unittests	86.94% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[henrydingliu]

equating outlier with drop_high=True is not theoretically sound. can you pick any actual outlier to drop (may involve displaying and inspecting the ldf) or use a different rationale for using drop_high=True?

[priyam0k]

Good catch. You're right that drop_high=True is not an outlier test, so that wording was misleading. I'll rephrase it as a sensitivity check on the column maxima and push the fix shortly.

[henrydingliu]

good to have international representation on this package