SYNC: add inventory_q.md from lecture-python.myst

[mmcky]

Summary

• Adds lectures/inventory_q.md, mirrored from QuantEcon/lecture-python.myst at upstream commit 78030a3a2 (latest commit touching this file, via PR Add lectures on Kihlstrom's paper, Ross Recovery Theorem and update lectures in probability theory lecture-python.myst#859; originally introduced in #827, with later updates in #830 and #859).
• First of three PRs implementing the plan agreed in TASK: reorder lectures #4 to introduce a new Markov Decision Processes section that groups discrete_dp with the upstream Q-learning lectures.
• File is not yet wired into _toc.yml — only_build_toc_files: true means it will not build until the TOC PR lands. This is intentional so the three PRs can land sequentially.

Follow-ups

• SYNC: rs_inventory_q.md (depends on this file existing, since it {doc}-references inventory_q).
• TOC: introduce Markov Decision Processes section (closes TASK: reorder lectures #4).
• Update DOC: Primary source mapping for all lecture files #7 to bump the lecture-python.myst mapping from 25 → 27 files.

Test plan

• CI passes (file is outside the TOC, so build should be unaffected).
• Diff against upstream is a clean copy with no unintended modifications.

🤖 Generated with Claude Code

[Copilot]

Pull request overview

This PR adds a new lecture page, lectures/inventory_q.md, introducing an inventory management model solved via dynamic programming and then via Q-learning, as part of the planned new Markov Decision Processes section.

Changes:

• Adds the new MyST/Jupytext-synced lecture inventory_q.md.
• Implements VFI and Q-learning code examples (Numba-accelerated) and accompanying figures/text.
• Introduces new cross-references/citations that will need to resolve cleanly once the page is included in the TOC.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.