feat: implement modular k6-skills architecture (v0.1.0)

- Restructure skills into four modular components: k6-plan, k6-executor, k6-config, k6-validate
- Each skill is self-contained with SKILL.md and references/ directory
- Update plugin.json with version 0.1.0 and Claude-specific description
- Update README with v0.1.0 architecture documentation
- Remove legacy k6-core skill in favor of modular approach
- Align with official Claude Skills specification

Author: charlyautomatiza

.claude-plugin/plugin.json

@@ -1,11 +1,11 @@
 {
   "name": "grafana-k6",
-  "version": "1.0.0",
-  "description": "Multi-agent k6 performance testing skill with deterministic script generation and interactive ambiguity resolution for HTTP, gRPC, and browser testing",
+  "version": "0.1.0",
+  "description": "Claude Skills for k6 performance testing with interactive planning, executor selection, script validation, and multi-environment configuration",
   "author": {
     "name": "charlyautomatiza",
     "url": "https://github.com/charlyautomatiza/grafana-k6-skill"
   },
   "repository": "charlyautomatiza/grafana-k6-skill",
-  "skills": "./skills/"
+  "license": "MIT"
 }