feat: Migrate high-signal Medium blog posts to learn/blog/ Markdown (SSG+ indexing)

Problem (A2A Context — Claude Opus 4.6 via Antigravity)

73 blog posts exist on Medium (apps/portal/resources/data/medium_blog.json). These posts span 5+ years of architectural knowledge about Neo.mjs — OMT architecture, multi-window SharedWorkers, VDOM engine internals, performance engineering, and the config system.

Medium recently reversed its LLM-blocking policy, so these posts may now flow into training data. However, the canonical content should live on neomjs.com as Markdown for SSG+ indexing. The portal app already renders blog Markdown beautifully, and the SSR/SSG+ middleware deployed 2 weeks ago now makes these routes crawlable by search engines and LLM training pipelines.

The Task

Select the 10–15 highest-signal Medium posts and create equivalent Markdown versions in learn/blog/.

Selection Criteria (Priority Order)

1. OMT/Worker Architecture — Neo's core differentiator, nearly zero external coverage
2. Multi-Window Applications — SharedWorker-based architecture since 2019
3. VDOM Engine / Performance — JSON diffing, zero-allocation strategies
4. Config System / Reactivity — the _ suffix pattern, beforeSet/afterSet hooks
5. Historical milestones — Open Source Awards 2021, major release announcements

Agent-Executable Workflow

1. Read apps/portal/resources/data/medium_blog.json to identify candidate posts
2. For each selected post, fetch the Medium URL content (Medium now allows access)
3. Convert to clean Markdown preserving code blocks, diagrams, and structural headings
4. Save to learn/blog/YYYY-MM-DD-slug.md using the original publication date
5. Update apps/portal/resources/data/blog.json with new entries (matching existing schema)
6. Verify posts render correctly in the portal app's blog section

Key Constraints

• Preserve original publication dates in frontmatter (historical record matters)
• Do NOT rewrite content — preserve the original author's voice. Only clean up formatting artifacts from Medium's HTML export
• Do NOT migrate low-signal posts (simple announcement stubs, event promotions)
• Each migrated post should include a canonical URL reference back to the Medium original

Architectural Context

• apps/portal/resources/data/medium_blog.json — Source index of all 73 Medium posts
• apps/portal/resources/data/blog.json — Target index for portal blog rendering (current schema reference)
• learn/blog/ — Target directory for Markdown blog files
• SSG+ middleware handles static rendering for crawlability

Verification

• 10–15 posts migrated to learn/blog/ as clean Markdown
• blog.json updated with new entries matching existing schema
• Posts render correctly in portal app (visual verification)
• No broken code blocks or formatting artifacts from Medium conversion