feat: reorganise bash scripts and add comprehensive documentation (#6)

## Summary

This PR completes the bash script restructuring and adds comprehensive documentation across all script categories.

### Bash Restructuring

- Moved cohort shell scripts (`remove_github_members.sh`, `update_github_teams.sh`) from `python/cohort/` into a new `bash/cohort/` directory
- Moved existing bash utilities (`add-keys-to-git.sh`, `fix-yubikey-perms.sh`, `list-yubikey-ssh-keys.sh`) into a new `bash/yubikey/` subdirectory
- Updated `run.sh` to support **Bash** as a third language option alongside TypeScript and Python
  - Bash scripts are run directly (no 1Password secret injection needed)
  - Category discovery and script listing works the same as for TS/Python
  - Removed dead "Root Scripts" logic that was no longer needed

### Documentation

Added `README.md` files for all script categories that were missing them:

- `bash/cohort/README.md` — cohort GitHub team management scripts
- `bash/yubikey/README.md` — YubiKey SSH key and permission utilities
- `typescript/src/crowdin/README.md` — Crowdin translation management scripts
- `typescript/src/discord/README.md` — Discord bot utility scripts
- `typescript/src/discourse/README.md` — Discourse forum management scripts
- `typescript/src/gitea/README.md` — Gitea bulk repository operation scripts
- `typescript/src/github/README.md` — GitHub API interaction scripts
- `typescript/src/music/README.md` — Music file metadata tools
- `typescript/src/s3/README.md` — S3-compatible object storage scripts
- `typescript/src/security/README.md` — Security analysis and reporting scripts
- `python/cohort/README.md` — Updated to remove moved shell scripts, fix usage commands

Also updated project-level docs:

- **`README.md`** — Corrected project structure, fixed running instructions (removed references to non-existent `make run-ts`/`make run-py` targets), added Bash prerequisites
- **`CLAUDE.md`** — Updated project overview, structure, development standards, and script-adding guides to reflect the current state of the project

✨ This PR was created with help from Hikari~ 🌸

Co-authored-by: Naomi Carrigan <commits@nhcarrigan.com>
Reviewed-on: #6
Co-authored-by: Hikari <hikari@nhcarrigan.com>
Co-committed-by: Hikari <hikari@nhcarrigan.com>

python/cohort/evaluate_technical_proficiency.py

@@ -1,8 +1,27 @@
+"""Evaluate the technical proficiency of cohort applicants using their GitHub profiles.
+
+Fetches each applicant's public GitHub repositories and scores their proficiency as
+Beginner, Intermediate, or Advanced based on language variety, repo count, commit
+activity, and presence of certain technologies.
+
+Data files (place in data/):
+  - applicants_to_evaluate.json  List of applicants with GitHub usernames
+
+Outputs (written to data/):
+  - proficiency_evaluations.json  Proficiency scores and tech stacks per applicant
+
+Env vars:
+  - None (uses public GitHub API; may be rate-limited without authentication)
+"""
+
 import json
 import re
 import time
 import urllib.error
 import urllib.request
+from pathlib import Path
+
+DATA_DIR = Path(__file__).parent.parent.parent / "data"
 
 # GitHub API (no auth needed for public repos, but rate limited)
 GITHUB_API = "https://api.github.com"
@@ -234,7 +253,7 @@ def evaluate_applicant(applicant: dict, index: int, total: int) -> dict:
 
 def main():
     # Load applicants
-    with open("applicants_to_evaluate.json") as f:
+    with open(DATA_DIR / "applicants_to_evaluate.json") as f:
         applicants = json.load(f)
 
     print(f"Evaluating {len(applicants)} applicants...\n")
@@ -249,7 +268,7 @@ def main():
             print(f"  Progress: {i + 1}/{len(applicants)} complete")
 
     # Save results
-    with open("proficiency_evaluations.json", "w") as f:
+    with open(DATA_DIR / "proficiency_evaluations.json", "w") as f:
         json.dump(evaluations, f, indent=2)
 
     # Summary