fix: armed trigger names now get watcher scoring (blank reasoning panels)

Four of ten armed triggers had NO criteria events ever — the trigger book
arms from the live gapper scan while the watcher scores only the
discovery watchlist, so scanner-only names (TCX/ZTG class) showed an
empty reasoning panel with nothing behind it.

- arm pass injects every complete armed candidate into the watchlist
  provider's extra_symbols (its standing manual-add hook)
- provider returns injected symbols FIRST: the watcher truncates at
  max_symbols, so extras appended after a full research list were
  silently dropped — 'always include manually injected symbols' was
  untrue exactly when the list was full

60 watcher/watchlist/trigger tests pass.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

research/ingestion/watcher_task.py

@@ -55,10 +55,29 @@ class ResearchWatchlistProvider:
             price_max=self.price_max,
             limit=self.limit,
         )
-        seen = set()
+        # Injected symbols FIRST: the watcher truncates at max_symbols, so
+        # extras appended after a full research list were silently dropped —
+        # exactly the armed-trigger names the reasoning panel needs scored.
         candidates = []
+        seen = set()
+        for symbol in sorted(self.extra_symbols):
+            seen.add(symbol)
+            candidates.append(
+                WatchCandidate(
+                    symbol=symbol,
+                    previous_close=research_query.query_previous_close(
+                        self.con, symbol, session_date
+                    ),
+                    avg_daily_volume=research_query.query_avg_daily_volume(
+                        self.con, symbol, session_date
+                    ),
+                    source="manual",
+                )
+            )
         for row in rows:
             symbol = row["symbol"]
+            if symbol in seen:
+                continue
             seen.add(symbol)
             candidates.append(
                 WatchCandidate(
@@ -73,20 +92,6 @@ class ResearchWatchlistProvider:
                     source="research",
                 )
             )
-        # always include manually injected symbols
-        for symbol in sorted(self.extra_symbols - seen):
-            candidates.append(
-                WatchCandidate(
-                    symbol=symbol,
-                    previous_close=research_query.query_previous_close(
-                        self.con, symbol, session_date
-                    ),
-                    avg_daily_volume=research_query.query_avg_daily_volume(
-                        self.con, symbol, session_date
-                    ),
-                    source="manual",
-                )
-            )
         return candidates
 
     def get_bars(self, symbol: str, session_date: date) -> pd.DataFrame:

run_live_paper.py

@@ -1016,6 +1016,17 @@ def main(argv: list[str] | None = None) -> int:
         # catalyst-backed gappers float to the top of the armed set
         candidates.sort(key=lambda c: c.get("_score", 0.0), reverse=True)
         book.arm(candidates)
+        # Armed names must also get watcher scoring: the trigger book arms from
+        # the live gapper scan, so names discovery never surfaced carried NO
+        # criteria trail — 2026-08-01 four of ten armed rows had a blank
+        # reasoning panel. extra_symbols is the provider's standing hook.
+        try:
+            _prov = rt["watcher"].provider
+            for c in candidates:
+                if c.get("symbol") and c.get("complete"):
+                    _prov.add_symbol(c["symbol"])
+        except Exception:  # noqa: BLE001 — observability aid, never blocks arming
+            pass
         armed = sum(1 for t in book.triggers.values() if t.state == "armed")
         n_news = sum(1 for c in candidates if c.get("catalyst"))
         return f"tracking={len(book.triggers)} armed={armed}" + (f" news={n_news}" if n_news else "")