v0.1.2

• Set the package version to 0.1.2 in pyproject.toml and __version__.
• Added docs/BUSINESS_VALUE.md and a README section that links to it. The workshop has not been delivered. Delivery cost 4,200 USD, rates, and ROI rows are [EST], with the incident term [SYN]. Docs only.
• Expanded the README: a three-line Start here, the lab agenda, learning outcomes, exercise architecture, per-lab commands, and a "What this does NOT prove" section. Docs only.
• Re-measured on 2026-10-22, package 0.1.1, Python 3.12.3, main commit 9d10f1d4d4dc8045ccb85749204a926bc2dba912: python -m pytest --cov=workshop --cov-report=term-missing --cov-fail-under=80 printed 48 passed and 92.79% total coverage (1456 statements, 105 missed). The same day: eval 6 of 6, checker vs author reference labels 5/6, demo SIMULATED cost 0.00012000 and SIMULATED latency 15 ms, solution modules 1–8 fail=0, secretscan findings=0, ruff clean, mypy clean on 79 files, eval-regression exit 1, llm-judge exit 2 without configuration. These counts match the 2026-10-18 run on 496a705e1a21a1d533b57be2160d93e29496d233. This docs change does not edit src/.
• Bumped actions/setup-python from 5 to 7.
• Added a three-line TL;DR under the README title and badges. Docs only.
• Bumped actions/checkout from 4 to 7.
• Added docs/demo/: a narrated recording of a real offline run, with English and Spanish subtitles and a number trace. Docs only.
• Added an Authorship section to the README. Docs only.
• Docs only. Security describes a teaching repo and points vulnerability reports at GitHub private vulnerability reporting.
• Re-measured on 2026-10-18, package 0.1.1, Python 3.12.3, main commit 496a705e1a21a1d533b57be2160d93e29496d233: eval 6 of 6, checker vs author reference labels 5/6, pytest 48 passed at 92.79% coverage, demo simulated cost 0.00012000 USD and simulated latency 15 ms, solution modules 1–8 fail=0, secretscan findings=0, ruff clean, mypy clean on 79 files, eval-regression exit 1, llm-judge exit 2 without configuration. The original workshop is PR #1, squash c590b91 on main. A live LLM-judge call was not re-measured.