Add LXX Morphology downloader and README chat-question template

[editor77]

Motivation

• Provide a small utility to fetch the LXX Morphology corpus from the UPenn mirror so the repository can populate a local *.mlxx corpus.
• Restrict downloads to MLXX files by default to avoid pulling unrelated mirror content and add a configurable User-Agent to increase compatibility with servers.
• Add a short chat-style question template to README.md to standardize requests for analysis runs against the local *.mlxx files.

Description

• Add scripts/download_lxxmorph.py, a recursive downloader that parses HTML/gopher-style listings, follows links, creates output directories, and saves files idempotently.
• Implement an --include argument (default \.mlxx$) to filter filenames and a --user-agent argument (default Mozilla/5.0 (compatible; LXXMorphDownloader/1.0)) which is threaded into requests via fetch_bytes.
• Ensure existing files are skipped, links are normalized with urllib.parse.urljoin, and directories are created with ensure_directory.
• Update README.md with usage notes and add a Chat question template example for asking analysis requests against the local .mlxx corpus and associated manual.

Testing

• No automated tests were run for this change.
• No network-based validation was executed as part of this rollout, so remote download behavior was not exercised by automated checks.

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d26e317163

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Prevent crawling outside the requested base URL

If the directory listing includes a parent link such as ../ (common in Apache-style indexes), parse_links will return it and the recursion will follow it because there is no guard that link stays under the original base_url. That means a run starting at .../lxxmorph/ can walk up to .../biblical/ and begin downloading unrelated directories (still matching *.mlxx), which is both unexpected and potentially huge. Consider rejecting links that are not prefixed by the initial base or explicitly skipping .. links before recursing.

Useful? React with 👍 / 👎.