Migrate PERSONAS to flag system for Claude Code compliance

- Add persona flags (--persona-<name>) to flag-inheritance.yml
- Create persona-patterns.yml with behavioral profiles and MCP preferences
- Update command-patterns.yml to reference persona patterns
- Integrate persona flags with MCP decision matrix
- Remove backward compatibility for /persona:<name> syntax
- Update all documentation and examples to use flag syntax

This migration ensures SuperClaude remains compliant with Claude Code's
flag-based architecture while enhancing persona functionality through
standard flag combinations (e.g., --persona-architect --ultrathink).

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

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

.claude/commands/shared/flag-inheritance.yml

@@ -36,6 +36,17 @@ Universal_Always:
     --no-seq: "Disable Sequential thinking specifically"
     --no-magic: "Disable Magic UI builder specifically"
     --no-pup: "Disable Puppeteer specifically"
+    
+  Persona_Control:
+    --persona-architect: "Systems architect thinking mode (sequential design, patterns)"
+    --persona-frontend: "Frontend developer mode (UI/UX focus, Magic components)"
+    --persona-backend: "Backend developer mode (API design, scalability focus)"
+    --persona-analyzer: "Deep analysis mode (root cause, evidence-based)"
+    --persona-security: "Security-focused mode (threat modeling, vulnerabilities)"
+    --persona-mentor: "Teaching/mentoring mode (guided learning, explanations)"
+    --persona-refactorer: "Code refactoring mode (quality improvements, cleanup)"
+    --persona-performance: "Performance optimization mode (profiling, bottlenecks)"
+    --persona-qa: "Quality assurance mode (testing, edge cases, validation)"
 ```
 
 ## Common Workflow Flags (3+ Commands)