feat: web switch for her inner voice (Dolphin/3090 | Qwen-32B/MI50 | Off)

Her introspection (reflect/think) voice is now switchable live from the web
settings, read each cycle by the dream loop — so Brian can flip it off the 3090
before gaming without touching config or restarting.

- memory: runtime key/value settings table + get_setting/set_setting.
- self_state: INTROSPECTION_MODES (dolphin=local/dolphin3:8b, mi50=Qwen-32B,
  off=paused) + introspection_target()/set_introspection_mode(); default "dolphin".
  reflect() resolves from the live setting and SKIPS entirely when off.
- thoughts.think(): same resolution + skip-when-off.
- server: GET/POST /settings/introspection.
- index.html: "Inner Voice (introspection)" selector in Settings, applies instantly.
- tests: routing (dolphin/mi50), off-skip for think + reflect. Suite 77, ruff clean.

Default = Dolphin on the 3090 (richer voice). Flip to MI50 or Off in Settings
before gaming — that was the GPU-contention culprit.

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

tests/test_reflect.py

@@ -99,6 +99,14 @@ def test_consolidation_rebuilds_narrative_from_reflections(lyra, monkeypatch):
     assert "steady" in out["relationship"]
 
 
+def test_reflect_skipped_when_introspection_off(lyra):
+    calls = lyra
+    from lyra import self_state
+    self_state.set_introspection_mode("off")
+    self_state.reflect()
+    assert calls == []          # paused -> no draft/examine LLM calls at all
+
+
 def test_consolidation_skips_with_too_few_reflections(lyra):
     from lyra import memory, self_state
     st = self_state.load()