feat: integrate image-based activities into Language Puzzles

[Gautham-2304]

Summary

Integrates six image-based language learning activities into the existing Language Puzzles module.

What's included

• Integrated six image-based activities into the Language Puzzles module:
  • Identify the Word (MCQ)
  • Fill in the Blank
  • Odd One Out
  • Word Pairing (Matching)
  • Guided Description
  • Free Description
• Added image-based learning content and assets
• Added Drishya exercise components
• Added Drishya content data and backend API
• Integrated the activities into LanguageDashboard
• Added API proxy configuration for Drishya
• Added the required backend route

Validation

• Frontend production build succeeds with npm run build
• server/routes/drishya.js passes node --check
• git diff --cached --check passes

Notes

The Drishya activities are now presented as part of the Language Puzzles module rather than as a separate user-facing lab.

Backend runtime testing is currently blocked by an existing syntax error in server/auth.js on the upstream main branch. server/auth.js was not modified as part of this contribution.