feat: implement tweaks plan - modals, smart shell, spinner, /models, debug log, skills

Phase 1: Permission modal dialog, session resume modal, HistoryInput with
up/down arrow cycling, remove "You:" echo from chat log, LLM client cleanup
on unmount.

Phase 2: Smart shell auto-approve using allowed/denied command lists from
ToolsConfig, animated thinking spinner with live token count in status bar.

Phase 3: /models slash command (list + switch), CLI directory positional
argument, JSONL debug logger with rotation.

Phase 4: Skills system with SkillsManager, load_skill LLM tool, /skills
listing, skill invocation via slash commands, system prompt integration.

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

app/services/llm.py

@@ -60,6 +60,25 @@ class LLMClient:
             timeout=httpx.Timeout(config.timeout, connect=10.0),
         )
 
+    async def list_models(self) -> list[dict[str, str]]:
+        """Query Ollama /api/tags for available models.
+
+        Returns:
+            List of dicts with 'name' and 'size' keys.
+
+        Raises:
+            LLMConnectionError: If the endpoint is unreachable.
+        """
+        try:
+            response = await self._client.get("/api/tags")
+            data = response.json()
+            return [
+                {"name": m.get("name", ""), "size": str(m.get("size", ""))}
+                for m in data.get("models", [])
+            ]
+        except (httpx.HTTPError, httpx.TimeoutException) as e:
+            raise LLMConnectionError(f"Failed to list models: {e}") from e
+
     async def preflight_check(self) -> None:
         """Verify the endpoint is reachable and the configured model is available.