feat(i18n): canonical strings, data extraction and validation (phase 2)

Implement Phase 2 of i18n spec:
- Extract UI interface strings to messages/pt.json across 15 namespaces
- Extract canonical data files to data/pt/ (series, frameworks, conexoes)
- Create lib/data/pt/ modules (glossario, diagnostico, componentes, topicos)
- Implement lib/data/index.js async data loaders by locale
- Add automated check:i18n script to validate key parity and enforce no-book rules

Co-Authored-By: Antigravity <noreply@google.com>

package.json

@@ -8,7 +8,8 @@
     "start": "next start",
     "lint": "eslint",
     "audit": "node scripts/auditar-paginas.mjs",
-    "check": "npm run lint && npm run auditar-temas && npm run audit",
+    "check:i18n": "node scripts/check-i18n.mjs",
+    "check": "npm run lint && npm run auditar-temas && npm run audit && npm run check:i18n",
     "favicons": "node scripts/generate-favicons.mjs",
     "auditar-temas": "node scripts/auditar-temas.mjs",
     "slides-pdf": "node scripts/slides-pdf.mjs"