Clean up: remove obsolete files, update docs and docstrings

Removed 20 files: old test scripts, debug tools, duplicate docs,
generated JSON, old PDF generator, launcher scripts.

Kept:
  disco.py        — main tool (scrape HAR + generate PDF)
  scraper.py      — reference site scraper (HTML + Selenium/Brave)
  requirements.txt
  *.har           — browser capture with API data

Updated:
  README.md       — rewritten to reflect current tool and usage
  .gitignore      — simplified
  scraper.py      — module/class/method docstrings updated to clarify
                    this is a reference implementation, disco.py is primary

scraper.py

@@ -1,7 +1,20 @@
 #!/usr/bin/env python3
 """
-Pokemon Discovery - TCG Product Scraper for Dollar General
-Scrapes product information and saves to JSON for PDF generation
+Pokemon Discovery — Site Scraper (Reference)
+
+HTML + Selenium/Brave scraper for Dollar General product pages.
+Kept as a reference implementation. The primary tool is disco.py,
+which reads product data from a HAR capture instead of scraping live.
+
+This scraper can:
+  - Fetch individual product pages and extract title, SKU, price, stock
+  - Attempt to find product links from the category page (limited by
+    dynamic JS loading — products are injected via API after page load)
+  - Fall back to Brave browser via Selenium for JS-rendered content
+
+Usage:
+    python scraper.py                  # Attempt full category scrape
+    # Or import and use PokemonTCGScraper class directly for individual pages
 """
 
 import json
@@ -28,6 +41,14 @@ except ImportError:
     print("Selenium not available, using requests only (install selenium for Brave browser support)")
 
 class PokemonTCGScraper:
+    """HTML/Selenium scraper for Dollar General Pokemon product pages.
+
+    Can extract product details (title, SKU, price, stock) from individual
+    product page URLs. Category-level scraping is limited because Dollar
+    General loads products dynamically via a JS API call after page load.
+    See disco.py for the HAR-based approach that bypasses this limitation.
+    """
+
     def __init__(self):
         self.base_url = "https://www.dollargeneral.com"
         self.search_url = "https://www.dollargeneral.com/c/toys/pokemon?q=&soldAtStore=true"
@@ -300,9 +321,10 @@ class PokemonTCGScraper:
         return has_pokemon and has_tcg
     
     def try_api_scraping(self):
-        """
-        Try to scrape products using the discovered API endpoint
-        This method contains the exact API call found via HAR analysis
+        """Stub for API-based scraping (requires auth token).
+
+        Documents the discovered API endpoint and request format.
+        Not functional — use disco.py with a HAR file instead.
         """
         print("🔬 Attempting API-based scraping...")
         print("   Endpoint: https://dggo.dollargeneral.com/omni/api/v2/category/search/provider")