Extraction: model in config, chatter gate, geocode cache (server-26#298, server-26#104, server-26#88)

EXTRACTION_MODEL / EMBEDDING_MODEL replace the gpt-4o-mini and
text-embedding-3-small literals. gate_chatter (classify_chatter with roll_call
narrowed to whole-transcript roll calls) skips correction, extraction,
embedding and geocoding for housekeeping, writes skip_reason=non_event_chatter
and still runs the free clearance/plate backstops; upload.py's no-scene
fallback admits it like transcript_too_short. Ships OFF (CHATTER_GATE_ENABLED
=false): a gated call has no units, so it takes the no-fit-check thin attach,
and #127 requires a clean live shadow window first.
In-process TTL cache (24h, negatives 1h, 5000 entries, LRU) in front of the
Geocoding API, keyed by address+bounds; holds coordinates only, inside the
Maps 30-day lat/lng caching limit; errors and key/quota statuses not cached.
#104: the no-scene fallback passes the call doc's severity.
#88: resolved/reassignment parsed by _llm_bool, not bool().

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

drb-c2-core/tests/test_intelligence_llm_bool.py

@@ -0,0 +1,14 @@
+"""server-26#88 — resolved/reassignment are parsed, not bool()-coerced."""
+import pytest
+
+from app.internal.intelligence import _llm_bool
+
+
+@pytest.mark.parametrize("value,expected", [
+    (True, True), (False, False),
+    ("true", True), ("TRUE", True), (" Yes ", True), ("yes", True),
+    ("false", False), ("False", False), ("no", False), ("NO", False),
+    ("", False), ("maybe", False), (None, False), (1, False), (0, False), ([], False),
+])
+def test_llm_bool(value, expected):
+    assert _llm_bool(value) is expected