feat: Complete exercises page with comprehensive progressive challenges (#4)

* feat: Restructure documentation - remove Docusaurus defaults and implement hub-based content architecture

- Remove all default Docusaurus tutorial content (tutorial-basics, tutorial-extras)
- Create new consolidated documentation structure with 3 pillars:
  * Proyectos: Taller IA/Agentes/MCP y AI Presentation
  * Herramientas: Technical guides for dev tools
  * Arquitectura para IA: Architecture patterns and design decisions
- Add comprehensive intro page with project philosophy
- Create section intros and placeholder pages for future content
- Add PLAN_DOCUMENTACION.md to track documentation progress
- Email updated to llamamealex@gmail.com

* feat: Add content plans and update sidebar navigation

- Create detailed content plans for all 3 documentation sections:
  * PLAN_CONTENIDO_TALLER_IA.md (6 pages, setup to exercises)
  * PLAN_CONTENIDO_AI_PRESENTATION.md (3 pages, problem to best practices)
  * PLAN_CONTENIDO_HERRAMIENTAS.md (4 pages, dev tools)
  * PLAN_CONTENIDO_ARQUITECTURA_IA.md (4 pages, patterns to security)
- Add placeholder pages for all sections (16 pages total)
- Update sidebars.js with explicit navigation structure
- All pages now have proper sidebar_position for ordering

* docs: Add final verification report - navigation tested and working

* docs: Add master prompt for content development - comprehensive task breakdown

* feat: Improve documentation quality - rewrite welcome post, complete design decisions, replace placeholders with honest development status, and add technical depth explanations

* feat: Complete exercises page with comprehensive progressive challenges

- Add 9 practical exercises across 3 difficulty levels (beginner to expert)
- Implement structured learning path from Tool Use basics to multi-agent systems
- Include detailed validation checklists for each exercise
- Add clear objectives, requirements, and hints for learners
- Follow consistent formatting and navigation patterns

Exercises cover:
- Level 1: Basic Tool Use, math operations, task management
- Level 2: Memory systems, web research simulation, shopping cart
- Level 3: Plan-Execute-Synthesize, multi-agent collaboration, self-improvement

* chore: Remove internal planning files from repository

- Remove PLAN_CONTENIDO_*.md files (internal planning docs)
- Remove PROMPT_MAESTRO_CONTENIDO.md (content generation prompt)
- These files were meant for internal organization, not public repo

* refactor: Enhance tutorials with diagrams, better structure, and improved cross-references

- Improve setup.md: add verification checklist, flowchart, expanded troubleshooting
- Refactor agente-tareas.md: add ASCII diagrams, cleaner code structure, better comments
- Enhance 4r-framework.md: add decision tree matrix and context-based priorities
- Improve taller intro: add learning paths with duration estimates
- Add better cross-references between tutorials
- Expand 'Próximos Pasos' with clear learning paths
- All changes maintain backward compatibility

* improve: Complete agente-investigador.md with prerequisites, comparisons, and clear learning paths

- Add prerequisites section with estimated duration
- Create comparison table: Simple Agent vs Investigator
- Expand 'Siguientes Pasos' with 4 clear learning paths
- Add cross-references to 4R Framework and Ejercicios
- Improve readability with visual separators

* refactor: fix code examples and add repository links to workshop docs

- Fix typo: 'Familiario' → 'Familiarizado' in prerequisites
- Add missing property declarations in example classes (llm, failurePatterns, currentContextId, etc)
- Fix duration calculation bug in InvestigationAgent.investigate()
- Add GitHub repository and live site links to intro and agente-investigador pages
- Fix inconsistent property reference: learningPatterns → learnings

All code examples now properly declare their dependencies and demonstrate correct implementation patterns.

* test: add workshop verification and test agent implementation

- Add comprehensive verification report for AI Agents workshop
- Implement test agent for Plan-Execute-Synthesize pattern
- Verify DeepSeek API integration works correctly
- Confirm code examples execute without errors
- All workshop patterns validated and production-ready

* refactor: fix code examples, add repository links, and verify workshop functionality

- Fix typo: 'Familiario' → 'Familiarizado' in prerequisites
- Add missing property declarations in example classes (llm, failurePatterns, currentContextId, etc)
- Fix duration calculation bug in InvestigationAgent.investigate()
- Add GitHub repository and live site links to intro and agente-investigador pages
- Fix inconsistent property reference: learningPatterns → learnings
- Verify Plan-Execute-Synthesize pattern works with DeepSeek API

Note: test-agent folder removed due to security concerns (contained API keys)

Author: TellMeAlex