Upgrade intent model to qwen2.5:7b, simplify intent prompt rules

config.py:
- Default OLLAMA_INTENT_MODEL: qwen2.5:1.5b → qwen2.5:7b
- Startup will auto-pull and warm the new model on next container restart

intent.py:
- Replaced phrase-matching examples in search_web and research_topic rules
  with semantic descriptions. The 7B model doesn't need example phrases to
  understand intent — it can reason from the tool's purpose. Removes implied
  usage patterns that caused misclassifications on conversational phrasing
  (e.g. "I've been thinking about buying shirts, can you research this?").
- research_topic rule now explicitly covers any subject regardless of phrasing,
  including shopping decisions, comparisons, how-things-work questions, etc.
- search_web rule clarified as "short summary, no note" vs research_topic's
  "comprehensive written reference"

The 1.5B model required prescriptive phrase examples to route correctly; the
7B model has sufficient language understanding to classify from semantic intent.
Expected improvement: ~1-2s intent calls (vs 0.4-9s for the 1.5B model which
sometimes timed out or misclassified longer/conversational messages).

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

src/fabledassistant/config.py

@@ -24,9 +24,9 @@ class Config:
     )
     OLLAMA_URL: str = os.environ.get("OLLAMA_URL", "http://localhost:11434")
     OLLAMA_MODEL: str = os.environ.get("OLLAMA_MODEL", "qwen3:latest")
-    # Optional dedicated model for intent classification (should be small/fast).
+    # Optional dedicated model for intent classification.
     # Falls back to OLLAMA_MODEL if not set. Can also be overridden per-user via settings.
-    OLLAMA_INTENT_MODEL: str = os.environ.get("OLLAMA_INTENT_MODEL", "qwen2.5:1.5b")
+    OLLAMA_INTENT_MODEL: str = os.environ.get("OLLAMA_INTENT_MODEL", "qwen2.5:7b")
     # KV cache context window for generation. Lower = less VRAM, less throughput impact.
     # 8192 is sufficient for most conversations; raise if you paste large documents.
     OLLAMA_NUM_CTX: int = int(os.environ.get("OLLAMA_NUM_CTX", "8192"))

src/fabledassistant/services/intent.py

@@ -100,14 +100,16 @@ Rules:
 - "read", "open", "show me", "what does X say", "display", "pull up" a specific note → use get_note with query=<note name>.
 - "list my notes", "show notes", "recent notes", "browse notes", "notes tagged X" → use list_notes (with optional q or tags).
 - "tag X with Y", "add tag Y to X", "untag Y from X", "remove tag Y from X" → use update_note with tags=[Y] and tag_mode="add" or "remove".
-- search_web: user wants a quick web search to answer a factual question
-  ("search for X", "look up X", "what is the latest version of X", "find X online",
-   "google X", "what is X" for quick factual answers — NOT when they want a comprehensive note)
-- research_topic: user wants to research a topic and create a comprehensive note from web sources
-  ("research X", "Research: X", "research X and make a note", "compile notes on X", "write a report on X",
-   "deep dive into X", "find everything about X", "comprehensive guide to X",
-   "research where to buy X", "research how to X", "research X and ship to me" — the topic
-   is everything after "Research:" or "research")
+- search_web: user wants a quick factual answer retrieved from the web, without creating a note.
+  Use for brief lookups where a short summary suffices (current version numbers, quick facts,
+  "what is X", "look up X"). Do NOT use when the user wants a detailed written reference.
+- research_topic: user wants a comprehensive, multi-section research note created from web sources.
+  Use whenever the user wants to deeply understand, learn about, or get a full written reference
+  on any subject — regardless of how they phrase it. The topic can be anything: technical subjects,
+  shopping decisions, comparisons, how things work, historical topics, etc.
+  The "topic" argument should capture the full subject matter of the request.
+  Prefer this over search_web when the user's request implies wanting thorough coverage rather than
+  a quick answer.
 - "ack": one short, natural sentence confirming the action (tool path only). Vary phrasing — do not always start with "Let me". Omit (null) for chat-only responses.
 - Do NOT wrap the JSON in markdown code fences."""