Blueprint v1.0.4: Implemented AI Context Optimization & Token Management

- core/utils.py: Added estimate_tokens(), truncate_to_tokens(), get_ai_cache(), set_ai_cache(), make_cache_key() utilities
- story/writer.py: Applied truncate_to_tokens() to prev_content (2000 tokens) and prev_sum (600 tokens) context injections
- story/editor.py: Applied truncate_to_tokens() to summary (1000t), last_chapter_text (800t), eval text (7500t), propagation contexts (2500t/3000t)
- web/routes/persona.py: Added MD5-keyed in-memory cache for persona analyze endpoint; truncated sample_text to 750 tokens
- ai/models.py: Added pre-dispatch payload size estimation with 30k-token warning threshold

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

core/utils.py

@@ -2,6 +2,7 @@ import os
 import json
 import datetime
 import time
+import hashlib
 from core import config
 import threading
 import re
@@ -19,6 +20,40 @@ _log_context = threading.local()
 # Cache for dynamic pricing from AI model selection
 PRICING_CACHE = {}
 
+# --- Token Estimation & Truncation Utilities ---
+
+def estimate_tokens(text):
+    """Estimate token count using a 4-chars-per-token heuristic (no external libs required)."""
+    if not text:
+        return 0
+    return max(1, len(text) // 4)
+
+def truncate_to_tokens(text, max_tokens):
+    """Truncate text to approximately max_tokens, keeping the most recent (tail) content."""
+    if not text:
+        return text
+    max_chars = max_tokens * 4
+    if len(text) <= max_chars:
+        return text
+    return text[-max_chars:]
+
+# --- In-Memory AI Response Cache ---
+
+_AI_CACHE = {}
+
+def get_ai_cache(key):
+    """Retrieve a cached AI response by key. Returns None if not cached."""
+    return _AI_CACHE.get(key)
+
+def set_ai_cache(key, value):
+    """Store an AI response in the in-memory cache keyed by a hash string."""
+    _AI_CACHE[key] = value
+
+def make_cache_key(prefix, *parts):
+    """Build a stable MD5 cache key from a prefix and variable string parts."""
+    raw = "|".join(str(p) for p in parts)
+    return f"{prefix}:{hashlib.md5(raw.encode('utf-8', errors='replace')).hexdigest()}"
+
 def set_log_file(filepath):
     _log_context.log_file = filepath