feat: initialize PKU-DFIIC data cleaning project

[QianbinDou]

Repurpose template repo as PKU Digital Finance Inclusion Index cleaning project:

• Add Python exploration script (01_explore.py): reads 4 Excel sheets, generates province/city/county/regionalism .dta files with full diagnostic output
• Add Stata cleaning scripts (02-04_clean_*.do): validation, labeling, xtset, and save for province (403 obs), city (4,369 obs), and county (26,090 obs)
• Add cleaned output files: PKU_DFIIC_province/city/county.dta
• Fix 3 Stata syntax issues found during first run:
  • assert does not accept message string arguments
  • r(unique_value) overwritten by duplicates list; save to local first
  • missing values treated as +inf by Stata; add !missing() to range checks
• Update CLAUDE.md and README.md for new project context
• Remove Beamer/LaTeX/R template scaffolding (Slides, Preambles, scripts, templates, etc.)
• Update .gitignore: exclude macOS resource forks (._*), data_temp/, data_logs/

Summary

Type

• Bug fix (fix/...)
• New feature: skill / agent / rule / hook (feat/...)
• Docs / guide / README (docs/...)
• Chore / cleanup (chore/...)

Test plan

• ./scripts/validate-setup.sh exits 0
• python3 scripts/quality_score.py <changed-files> ≥ 80
• /deep-audit finds no new inconsistencies
• Manually exercised the changed skill/agent/hook on a real file
• Updated both README.md and guide/workflow-guide.qmd if user-facing
• Skill counts agree across CLAUDE.md, README.md, docs/index.html, and the guide

Generality (for new skills/agents/rules)

• Works for ≥2 academic domains (not specific to one field)
• No hardcoded paths, machine-specific config, or institutional branding
• No project-specific examples in shared infrastructure

Notes for reviewer