feat(memory.search): optional minScore RRF threshold

Closes the long-standing backlog item from abfa5463. memory.search now
accepts an optional `minScore` parameter (Zod range 0..1) that drops hits
below the given Reciprocal Rank Fusion score. Default behavior is
unchanged — when minScore is unset, every fused result is returned, same
as today.

The original design memo suggested defaulting to 0.020, but that would
exclude valid pure-semantic matches (one ranker at rank 1 = 1/61 ≈
0.0164). Real-world Phase 2 testing surfaced exactly that case (the
"expose TLS" → HAProxy memory hit). Shipping unfiltered-by-default and
exposing the knob lets specific callers opt into stricter filtering
(e.g. ~0.025 to require two rankers to fire at rank 1) without
penalising legitimate semantic-only hits for the rest.

Tool description updated; per-source rank breakdown remains the primary
confidence signal for the model.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

packages/schemas/src/index.ts

@@ -90,6 +90,15 @@ export const MemorySearchInput = z.object({
   scope: MemoryScope.optional(),
   tags: z.array(z.string()).optional(),
   limit: z.number().int().min(1).max(50).default(10),
+  /**
+   * Minimum Reciprocal Rank Fusion score a result must clear to be
+   * returned. Useful for stricter "high-confidence only" filtering — set
+   * higher than 1/(60+1)≈0.0164 to exclude single-ranker-rank-1 matches
+   * (semantic-only hits with no FTS/tag corroboration), or to ~0.03 to
+   * require at least two rankers to fire at rank 1. Omit / set 0 for the
+   * unfiltered default.
+   */
+  minScore: z.number().min(0).max(1).optional(),
 });
 export type MemorySearchInput = z.infer<typeof MemorySearchInput>;