feat: split introspection backend from consolidation (trial Dolphin for her voice)

reflect()/think() can now run on a different model than memory consolidation:
INTROSPECTION_BACKEND / INTROSPECTION_MODEL (default to SUMMARY_BACKEND, so unset =
unchanged). Consolidation (summaries/profile/narrative) keeps the capable model;
her *voice* (reflections, thoughts) can run a steerable tune. dream.py lets
reflect()/think() self-resolve to the introspection backend; both now thread a
`model` override into llm.complete.

Trial live: introspection -> dolphin3:8b on the 3090; consolidation -> Qwen-32B
on the MI50. Suite 73 green, ruff clean.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

lyra/dream.py

@@ -110,13 +110,15 @@ def dream_cycle(backend: Backend | None = None, force: bool = False) -> dict:
 
     # --- curiosity: reflect and evolve the self, then advance the thought loop ---
     if force or drives["curiosity"] >= THRESHOLD:
-        self_state.reflect(backend=backend, source="dream")  # writes state + journal itself
+        # reflect()/think() self-resolve to the *introspection* backend (her voice),
+        # which can differ from the consolidation backend above — don't pass `backend`.
+        self_state.reflect(source="dream")  # writes state + journal itself
         actions.append("reflected")
         # Thinking, continued: advance one threaded train of thought. reflect()
         # just refreshed her self-state, so the thought is grounded in it. A bad
         # think pass shouldn't sink the cycle.
         try:
-            rep = thoughts.think(backend=backend, source="dream")
+            rep = thoughts.think(source="dream")
             actions.append(f"thought ({rep['mode']})" if rep else "thought (no parse)")
         except Exception as exc:
             logbus.log("error", "thought loop failed", error=str(exc)[:200])