feat(backend): implement MCP client infrastructure (#55)

Core MCP client implementation with comprehensive tooling:

**Services:**
- MCPClientManager: Main facade for all MCP operations
- MCPServerRegistry: Thread-safe singleton for server configs
- ConnectionPool: Connection pooling with auto-reconnection
- ToolRouter: Automatic tool routing with circuit breaker
- AsyncCircuitBreaker: Custom async-compatible circuit breaker

**Configuration:**
- YAML-based config with Pydantic models
- Environment variable expansion support
- Transport types: HTTP, SSE, STDIO

**API Endpoints:**
- GET /mcp/servers - List all MCP servers
- GET /mcp/servers/{name}/tools - List server tools
- GET /mcp/tools - List all tools from all servers
- GET /mcp/health - Health check all servers
- POST /mcp/call - Execute tool (admin only)
- GET /mcp/circuit-breakers - Circuit breaker status
- POST /mcp/circuit-breakers/{name}/reset - Reset circuit breaker
- POST /mcp/servers/{name}/reconnect - Force reconnection

**Testing:**
- 156 unit tests with comprehensive coverage
- Tests for all services, routes, and error handling
- Proper mocking and async test support

**Documentation:**
- MCP_CLIENT.md with usage examples
- Phase 2+ workflow documentation

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

docs/development/WORKFLOW.md

@@ -214,6 +214,71 @@ test(frontend): add unit tests for ProjectDashboard
 
 ---
 
+## Phase 2+ Implementation Workflow
+
+**For complex infrastructure issues (Phase 2 MCP, core systems), follow this rigorous process:**
+
+### 1. Branch Setup
+```bash
+# Create feature branch from dev
+git checkout dev && git pull
+git checkout -b feature/<issue-number>-<description>
+```
+
+### 2. Planning Phase
+- Read the issue thoroughly, understand ALL sub-tasks
+- Identify components and their dependencies
+- Determine if multi-agent parallel execution is appropriate
+- Create a detailed execution plan before writing any code
+
+### 3. Implementation with Continuous Testing
+**After EACH sub-task:**
+- [ ] Run unit tests for the component
+- [ ] Run integration tests if applicable
+- [ ] Verify type checking passes
+- [ ] Verify linting passes
+- [ ] Keep coverage high (aim for >90%)
+
+**Both modules must be tested:**
+- Backend: `IS_TEST=True uv run pytest` + E2E tests
+- Frontend: `npm test` + E2E tests (if applicable)
+
+### 4. Multi-Agent Review (MANDATORY before considering done)
+Before closing an issue, perform deep review from multiple angles:
+
+| Review Type | Focus Areas |
+|-------------|-------------|
+| **Code Quality** | Logic errors, edge cases, race conditions, error handling |
+| **Security** | OWASP Top 10, input validation, authentication, authorization |
+| **Performance** | N+1 queries, memory leaks, inefficient algorithms |
+| **Architecture** | Pattern adherence, separation of concerns, extensibility |
+| **Testing** | Coverage completeness, test quality, edge case coverage |
+| **Documentation** | Code comments, README, API docs, usage examples |
+
+**No stone unturned. No sloppy results. No unreviewed work.**
+
+### 5. Final Validation
+- [ ] All tests pass (unit, integration, E2E)
+- [ ] Type checking passes
+- [ ] Linting passes
+- [ ] Documentation updated
+- [ ] Coverage meets threshold
+- [ ] Issue checklist 100% complete
+- [ ] Multi-agent review passed
+
+### When to Use Parallel Agents
+Use multiple agents working in parallel when:
+- Sub-tasks are independent (no shared state/dependencies)
+- Different expertise areas (backend vs frontend)
+- Time-critical deliveries with clear boundaries
+
+Do NOT use parallel agents when:
+- Tasks share state or have dependencies
+- Sequential testing is required
+- Integration points need careful coordination
+
+---
+
 ## Quick Reference
 
 | Action | Command/Location |