Add web admin features + fix scraper & translator

Frontend changes:
- Add /admin dashboard for article management
- Add AdminButton component (Alt+Shift+A on articles)
- Add /api/admin/article API endpoints

Backend improvements:
- scraper_v2.py: Multi-layer fallback extraction (newspaper → trafilatura → readability)
- translator_v2.py: Better chunking, repetition detection, validation
- admin_tools.py: CLI admin commands
- test_scraper.py: Individual source testing

Docs:
- WEB-ADMIN-GUIDE.md: Web admin usage
- ADMIN-GUIDE.md: CLI admin usage
- SCRAPER-IMPROVEMENT-PLAN.md: Scraper fixes details
- TRANSLATION-FIX.md: Translation improvements
- ADMIN-FEATURES-SUMMARY.md: Implementation summary

Fixes:
- Article scraping from 0 → 96+ articles working
- Translation quality issues (repetition, truncation)
- Added 13 new RSS sources

backend/config.py

@@ -12,35 +12,19 @@ DATABASE_URL = os.getenv('DATABASE_URL', 'postgresql://localhost/burmddit')
 ANTHROPIC_API_KEY = os.getenv('ANTHROPIC_API_KEY')
 OPENAI_API_KEY = os.getenv('OPENAI_API_KEY')  # Optional, for embeddings
 
-# Scraping sources - 🔥 EXPANDED for more content!
+# Scraping sources - 🔥 V2 UPDATED with working sources!
 SOURCES = {
-    'medium': {
-        'enabled': True,
-        'tags': ['artificial-intelligence', 'machine-learning', 'chatgpt', 'ai-tools', 
-                'generative-ai', 'deeplearning', 'prompt-engineering', 'ai-news'],
-        'url_pattern': 'https://medium.com/tag/{tag}/latest',
-        'articles_per_tag': 15  # Increased from 10
-    },
+    # WORKING SOURCES (tested 2026-02-26)
     'techcrunch': {
         'enabled': True,
         'category': 'artificial-intelligence',
         'url': 'https://techcrunch.com/category/artificial-intelligence/feed/',
-        'articles_limit': 30  # Increased from 20
-    },
-    'venturebeat': {
-        'enabled': True,
-        'url': 'https://venturebeat.com/category/ai/feed/',
-        'articles_limit': 25  # Increased from 15
+        'articles_limit': 30
     },
     'mit_tech_review': {
         'enabled': True,
         'url': 'https://www.technologyreview.com/feed/',
         'filter_ai': True,
-        'articles_limit': 20  # Increased from 10
-    },
-    'theverge': {
-        'enabled': True,
-        'url': 'https://www.theverge.com/ai-artificial-intelligence/rss/index.xml',
         'articles_limit': 20
     },
     'wired_ai': {
@@ -48,13 +32,100 @@ SOURCES = {
         'url': 'https://www.wired.com/feed/tag/ai/latest/rss',
         'articles_limit': 15
     },
-    'arstechnica': {
+    
+    # NEW HIGH-QUALITY SOURCES (Priority Tier 1)
+    'openai_blog': {
         'enabled': True,
+        'url': 'https://openai.com/blog/rss/',
+        'articles_limit': 10
+    },
+    'huggingface': {
+        'enabled': True,
+        'url': 'https://huggingface.co/blog/feed.xml',
+        'articles_limit': 15
+    },
+    'google_ai': {
+        'enabled': True,
+        'url': 'http://googleaiblog.blogspot.com/atom.xml',
+        'articles_limit': 15
+    },
+    'marktechpost': {
+        'enabled': True,
+        'url': 'https://www.marktechpost.com/feed/',
+        'articles_limit': 25
+    },
+    'the_rundown_ai': {
+        'enabled': True,
+        'url': 'https://rss.beehiiv.com/feeds/2R3C6Bt5wj.xml',
+        'articles_limit': 10
+    },
+    'last_week_ai': {
+        'enabled': True,
+        'url': 'https://lastweekin.ai/feed',
+        'articles_limit': 10
+    },
+    'ai_news': {
+        'enabled': True,
+        'url': 'https://www.artificialintelligence-news.com/feed/rss/',
+        'articles_limit': 20
+    },
+    
+    # NEW SOURCES (Priority Tier 2)
+    'kdnuggets': {
+        'enabled': True,
+        'url': 'https://www.kdnuggets.com/feed',
+        'articles_limit': 20
+    },
+    'the_decoder': {
+        'enabled': True,
+        'url': 'https://the-decoder.com/feed/',
+        'articles_limit': 20
+    },
+    'ai_business': {
+        'enabled': True,
+        'url': 'https://aibusiness.com/rss.xml',
+        'articles_limit': 15
+    },
+    'unite_ai': {
+        'enabled': True,
+        'url': 'https://www.unite.ai/feed/',
+        'articles_limit': 15
+    },
+    'simonwillison': {
+        'enabled': True,
+        'url': 'https://simonwillison.net/atom/everything/',
+        'articles_limit': 10
+    },
+    'latent_space': {
+        'enabled': True,
+        'url': 'https://www.latent.space/feed',
+        'articles_limit': 10
+    },
+    
+    # BROKEN SOURCES (disabled temporarily)
+    'medium': {
+        'enabled': False,  # Scraping broken
+        'tags': ['artificial-intelligence', 'machine-learning', 'chatgpt'],
+        'url_pattern': 'https://medium.com/tag/{tag}/latest',
+        'articles_per_tag': 15
+    },
+    'venturebeat': {
+        'enabled': False,  # RSS feed empty
+        'url': 'https://venturebeat.com/category/ai/feed/',
+        'articles_limit': 25
+    },
+    'theverge': {
+        'enabled': False,  # RSS feed empty
+        'url': 'https://www.theverge.com/ai-artificial-intelligence/rss/index.xml',
+        'articles_limit': 20
+    },
+    'arstechnica': {
+        'enabled': False,  # Needs testing
         'url': 'https://arstechnica.com/tag/artificial-intelligence/feed/',
         'articles_limit': 15
     },
     'hackernews': {
-        'enabled': True,
+        'enabled': False,  # Needs testing
         'url': 'https://hnrss.org/newest?q=AI+OR+ChatGPT+OR+OpenAI',
         'articles_limit': 30
     }