Replace/Supplement Jodit WYSIWYG with Editor.js

[willgearty]

Add Editor.js as an opt-in block-based editing mode for QSD, with content stored as typed blocks instead of HTML. Existing Jodit/HTML mode stays as-is. Goal: drag-and-drop structured blocks (callout, image+caption, two-column, embed) rendered through ESP templates.

Why

QSD's editing UX is in a good place after recent work (Jodit, history viewer with restore, inline image upload, media library in flight). The remaining gap is structured content: anything beyond prose+image currently has to be hand-authored HTML, which causes design drift, makes content opaque to programmatic changes, and raises the floor for chapter admins.

Migration

• format defaults to "html". Existing QSDs untouched.
• Admins opt in per entry. HTML mode is supported indefinitely as a peer mode, not deprecated.
• "Convert to blocks" action wraps existing HTML in a single raw_html block as a starting point.

Alternatives considered

Wagtail (too heavy, routing conflicts), GrapesJS (outputs HTML not structured data, commercial pivot), BlockNote/Plate (React dependency), Tiptap (text- not block-oriented). Editor.js is the lightest option that actually delivers structured content.

(Claude helped draft this and I edited it way down)

[willgearty]

Related to #2885
Would likely resolve #841, #2746, #667
Would supersede #1270

[abhinavjha0239]

@willgearty I want to pick this.

[willgearty]

Just an idea for discussion right now (hence the label)

[rafiya618]

Do we see QSD evolving toward reusable structured content, or will most entries remain simple article-style content?

[willgearty]

I don't think QSD will ever be more than just simple article-style content.

[rafiya618]

That makes sense, if QSD is expected to remain primarily article style content, I think we can satisfy most of the proposed requirements within the existing Jodit workflow without introducing a second editor, storage format, and rendering pipeline.

e.g:

• callouts: add custom Jodit toolbar buttons that insert predefined callout templates (info, warning, success) with standardized classes.
• Image + Caption: use the existing image workflow and wrap inserted images in a figure/figcaption structure through a custom dialog or plugin.
• embedding: provide an embed dialog that inserts approved embed templates (YouTube, Vimeo, etc.) instead of requiring manual HTML.
• two-column layouts: add predefined layout templates/snippets that insert a controlled two-column HTML structure when needed.
• design consistency: enforce styling through shared CSS classes and templates rather than allowing arbitrary custom markup.
• media library integration: continue using the media library as the source of truth for image insertion and management.
• reusable content patterns: add snippets/templates for commonly used content structures so admins don't need to hand-author HTML.

that would preserve the current HTML based workflow while reducing the need for manual markup and maintaining consistent presentation across QSDs.

[github-actions]

⚠️ Sorry @krishna5379chaudhary-eng, this issue is labeled idea for discussion and is not currently available for assignment.