feat: add collector daemon and 5 vector search tools (106 total)

- collector.py: standalone daemon for 24/7 vector store population
  - 43 sources organized by domain with parallel fetching
  - --daemon, --interval, --sources CLI args
  - Dynamic import via _import_fetch_fn()
  - Entry point: intel-collector
- server.py: 5 new MCP tools (intel_semantic_search, intel_similar_events,
  intel_timeline, intel_vector_stats, intel_collect)
- pyproject.toml: vector optional dependency (qdrant-client + fastembed)
- README.md: updated architecture, vector store section, tool reference
- CLAUDE.md: updated tool count and architecture docs

pyproject.toml

@@ -35,6 +35,7 @@ dev = [
 world-intel-mcp = "world_intel_mcp.server:run"
 intel = "world_intel_mcp.cli:main"
 intel-dashboard = "world_intel_mcp.dashboard.app:run"
+intel-collector = "world_intel_mcp.collector:main"
 
 [build-system]
 requires = ["hatchling"]