SingleFile CLI 2.2.0

New features

Added --crawl-save-archive to save all the crawled pages into a single self-extracting ZIP file. The archive displays the pages with hash-based navigation, supports deep links to inner pages, and can be unzipped to browse the pages offline. It requires --compress-content
Added --crawl-save-archive-dedup to deduplicate identical resources shared between pages in the archive
Added --crawl-save-archive-mark-unarchived-links to mark links pointing to pages not saved in the archive
Added --crawl-save-archive-toc to store a table of contents page in the archive, accessible from the #sfz/?toc route

Crawl options like --crawl-no-parent or --crawl-max-depth now fail with an error when --crawl-links is not set. They were silently ignored before (#189 by @brasswood)
Clarified the syntax of --filename-replaced-character in the help message (#184 by @paraknell)
Digest template variables, e.g. {url-href-digest-sha-1}, now work when saving pages served over http: (single-file-core#5)

The image is now based on node:24-alpine and uses the Chromium version packaged by Alpine Linux. This fixes the "WebSocket is not available" error caused by the outdated Node.js version of the previous base image (#205 by @k1rinh)
The image is now built and published automatically when a release is published, for the linux/amd64 and linux/arm64 platforms, and tagged with the release version in addition to latest

Co-authored by Claude (Claude Code)