refactor: Remove wizards module to focus MemDocs on core competencies

MemDocs core mission is now clearly defined:
- Git-native memory storage for AI assistants
- Cross-conversation LLM memory (Claude Code, etc.)
- Enterprise audit-ready security (PHI/PII redaction)

The EmpathySyncWizard has been removed because:
- Wizards belong in domain-specific archives (healthcare, software dev, business)
- MemDocs should receive data, not orchestrate external services
- Cleaner separation: Empathy Framework owns wizard execution

What was removed:
- memdocs/wizards/__init__.py
- memdocs/wizards/empathy_sync.py (~400 lines)

What stays:
- empathy_adapter.py (transforms Empathy format → MemDocs storage)
- All core functionality intact
- 380 tests passing, 78% coverage maintained

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <[email protected]>

Author: 2 people

CONTRIBUTING.md

@@ -239,7 +239,7 @@ memdocs/
 │   ├── embeddings.py    # Vector embeddings
 │   ├── search.py        # Vector search
 │   ├── index.py         # Memory indexing
-│   └── wizards/         # High-level wizards
+│   └── empathy_adapter.py  # Empathy Framework integration
 ├── tests/
 │   ├── unit/            # Unit tests
 │   ├── integration/     # Integration tests

PRODUCTION_ROADMAP.md

@@ -29,12 +29,6 @@
   - Test error handling and edge cases
   - Integration test with Claude Desktop (optional)
 
-- [ ] **wizards/empathy_sync.py (0% → 75%)**:
-  - Mock Empathy API calls
-  - Test sync wizard end-to-end
-  - Test conflict resolution
-  - Test rollback on failure
-
 #### Medium Priority:
 - [ ] Increase extract.py: 80% → 90%
 - [ ] Increase symbol_extractor.py: 79% → 90%