feat(router,inference,config): pin narration to a configured model_id

[router.narration] model_id selects which provider handles narration turns
instead of generic capability routing — lets a small/fast model own narration
while the main model keeps CHAT/STEERING.

- NarrationSettings.modelId parsed from TOML, threaded via RouterConfig
  .narrationModelId into DefaultRouterFacade.narrate (3-arg route)
- DefaultInferenceRouter: exact-id match against healthy providers, with a
  post-select health check; any miss logs WARN and falls back to capability
  routing (never fails the narration turn)
- onUserInput unchanged — only narrate uses the pinned model (tested)

core/inference/build.gradle

@@ -8,5 +8,6 @@ dependencies {
     implementation(project(":core:events"))
     implementation(project(":core:context"))
     implementation(project(":core:artifacts"))
+    implementation "org.slf4j:slf4j-api:2.0.16"
 }
 tasks.named("koverVerify").configure { enabled = false }