fix: Phase 2 integrity review - v2.0.0 rewrite & bugfixes

Memory Consolidation Plugin (828 -> 465 lines):
- Replace SentenceTransformer with cat.embedder.embed_query() for vector consistency
- Fix per-user fact isolation: source=user_id instead of global
- Add duplicate fact detection (_is_duplicate_fact, score_threshold=0.85)
- Remove ~350 lines of dead async run_consolidation() code
- Remove duplicate declarative search in before_cat_sends_message
- Unify trivial patterns into TRIVIAL_PATTERNS frozenset
- Remove all sys.stderr.write debug logging
- Remove sentence-transformers from requirements.txt (no external deps)

Loguru Fix (cheshire-cat/cat/log.py):
- Patch Cat v1.6.2 loguru format to provide default extra fields
- Fixes KeyError: 'original_name' from third-party libs (fastembed)
- Mounted via docker-compose volume

Discord Bridge:
- Copy discord_bridge.py to cat-plugins/ (was empty directory)

Test Results (6/7 pass, 100% fact recall):
- 11 facts extracted, per-user isolation working
- Duplicate detection effective (+2 on 2nd run)
- 5/5 natural language recall queries correct

cat-plugins/discord_bridge/discord_bridge.py

@@ -0,0 +1,109 @@
+"""
+Discord Bridge Plugin for Cheshire Cat
+
+This plugin enriches Cat's memory system with Discord context:
+- Unified user identity across all servers and DMs
+- Guild/channel metadata for context tracking
+- Minimal filtering before storage (only skip obvious junk)
+- Marks memories as unconsolidated for nightly processing
+
+Phase 1 Implementation
+"""
+
+from cat.mad_hatter.decorators import hook
+from datetime import datetime
+import re
+
+
+@hook(priority=100)
+def before_cat_reads_message(user_message_json: dict, cat) -> dict:
+    """
+    Enrich incoming message with Discord metadata.
+    This runs BEFORE the message is processed.
+    """
+    # Extract Discord context from working memory or metadata
+    # These will be set by the Discord bot when calling the Cat API
+    guild_id = cat.working_memory.get('guild_id')
+    channel_id = cat.working_memory.get('channel_id')
+    
+    # Add to message metadata for later use
+    if 'metadata' not in user_message_json:
+        user_message_json['metadata'] = {}
+    
+    user_message_json['metadata']['guild_id'] = guild_id or 'dm'
+    user_message_json['metadata']['channel_id'] = channel_id
+    user_message_json['metadata']['timestamp'] = datetime.now().isoformat()
+    
+    return user_message_json
+
+
+@hook(priority=100)
+def before_cat_stores_episodic_memory(doc, cat):
+    """
+    Filter and enrich memories before storage.
+    
+    Phase 1: Minimal filtering
+    - Skip only obvious junk (1-2 char messages, pure reactions)
+    - Store everything else temporarily
+    - Mark as unconsolidated for nightly processing
+    """
+    message = doc.page_content.strip()
+    
+    # Skip only the most trivial messages
+    skip_patterns = [
+        r'^\w{1,2}$',  # 1-2 character messages: "k", "ok"
+        r'^(lol|lmao|haha|hehe|xd|rofl)$',  # Pure reactions
+        r'^:[\w_]+:$',  # Discord emoji only: ":smile:"
+    ]
+    
+    for pattern in skip_patterns:
+        if re.match(pattern, message.lower()):
+            print(f"🗑️  [Discord Bridge] Skipping trivial message: {message}")
+            return None  # Don't store at all
+    
+    # Add Discord metadata to memory
+    doc.metadata['consolidated'] = False  # Needs nightly processing
+    doc.metadata['stored_at'] = datetime.now().isoformat()
+    
+    # Get Discord context from working memory
+    guild_id = cat.working_memory.get('guild_id')
+    channel_id = cat.working_memory.get('channel_id')
+    
+    doc.metadata['guild_id'] = guild_id or 'dm'
+    doc.metadata['channel_id'] = channel_id
+    doc.metadata['source'] = cat.user_id  # CRITICAL: Cat filters episodic by source=user_id!
+    doc.metadata['discord_source'] = 'discord'  # Keep original value as separate field
+    
+    print(f"💾 [Discord Bridge] Storing memory (unconsolidated): {message[:50]}...")
+    print(f"   User: {cat.user_id}, Guild: {doc.metadata['guild_id']}, Channel: {channel_id}")
+    
+    return doc
+
+
+@hook(priority=50)
+def after_cat_recalls_memories(cat):
+    """
+    Log memory recall for debugging.
+    Access recalled memories via cat.working_memory.
+    """
+    import sys
+    sys.stderr.write("🧠 [Discord Bridge] after_cat_recalls_memories HOOK CALLED!\n")
+    sys.stderr.flush()
+    
+    # Get recalled memories from working memory
+    episodic_memories = cat.working_memory.get('episodic_memories', [])
+    declarative_memories = cat.working_memory.get('declarative_memories', [])
+    
+    if episodic_memories:
+        print(f"🧠 [Discord Bridge] Recalled {len(episodic_memories)} episodic memories for user {cat.user_id}")
+        # Show which guilds the memories are from
+        guilds = set()
+        for doc, score in episodic_memories:
+            guild = doc.metadata.get('guild_id', 'unknown')
+            guilds.add(guild)
+        print(f"   From guilds: {', '.join(guilds)}")
+
+
+# Plugin metadata
+__version__ = "1.0.0"
+__description__ = "Discord bridge with unified user identity and sleep consolidation support"

cat-plugins/memory_consolidation/requirements.txt

@@ -1 +0,0 @@
-sentence-transformers>=2.2.0