llm: switch Skald to agent-loop Model clients; drop llm-client (phase 1, D13)

The LLM call path now runs on the agent-loop crate's clients and trait:

- core-api: BuiltLlmClient.client is Arc<dyn agent_loop::model::Model>;
  chatbot.rs (ChatbotClient + wire types) deleted; APP_NAME re-exported
  from agent-loop
- providers (openai/anthropic/ollama/openrouter/requesty/declared) build
  OpenAiModel/AnthropicModel/OllamaModel with the model's wire id
- LoggingModel decorator (llm/logging.rs) replaces LoggingChatbotClient;
  per-request correlation (session/stack/user) travels in the new
  ModelRequest.log field, never sent to providers
- llm_call/llm_loop/compactor speak Model::complete + ModelResponse;
  retriability via Model::is_retriable (structured status, B6 rule now
  the crate's default); payload persistence reads RawMeta off
  ModelResponse/ModelError
- crates/llm-client and skald-core/src/chatbot deleted

Full workspace test suite green (incl. 162 skald-core + 32 agent-loop).

crates/core-api/src/provider.rs

@@ -3,7 +3,8 @@ use std::sync::Arc;
 use anyhow::Result;
 use async_trait::async_trait;
 
-use crate::chatbot::ChatbotClient;
+use agent_loop::model::Model;
+
 use crate::image_generate::{ImageGenerate, ImageGenerateModelRecord};
 use crate::tts::{TextToSpeech, TtsModelRecord, RemoteTtsModelInfo};
 use crate::transcribe::{Transcribe, TranscribeModelRecord, RemoteTranscribeModelInfo};
@@ -139,7 +140,8 @@ pub struct ProviderField {
 // ── BuiltLlmClient ────────────────────────────────────────────────────────────
 
 pub struct BuiltLlmClient {
-    pub client:       Arc<dyn ChatbotClient>,
+    /// A stateless `agent_loop` model client (blueprint D13).
+    pub client:       Arc<dyn Model>,
     pub prompt_cache: bool,
 }