Unified NL search Phase 2: /photos/search/unified endpoint

Composes the two existing engines (Path A orchestration):
- Translate NL -> StructuredQuery via local LLM, respecting LLM_BACKEND
  (resolve_backend(Local) -> ollama or llama-swap; no hybrid).
- Forward-geocode the place name into a gps circle.
- Structured filters (tags/EXIF/geo/date/media) build a candidate set of EXIF
  rows; CLIP ranks within it, joined by content_hash. Degenerate cases match
  existing behavior: semantic-only -> plain CLIP; filters-only -> date-sorted.
- Echoes the interpreted query (incl. resolved place) for editable client chips.

Refactor: extracted reusable cores from clip_search (score_photos, resolve_hits,
parse_library_scope, score_error_response) shared by both endpoints. Removed the
Phase 1 allow-until-wired attributes now that nl_query + geo are consumed.

fmt + clippy clean; 23 backend tests pass (7 geo, 12 nl_query, 4 unified).

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

src/ai/nl_query.rs

@@ -21,12 +21,6 @@
 //! `geo::forward_geocode`), and person filtering is deferred until a
 //! person→photos resolver exists.
 
-// Phase 1: this module is fully implemented and unit-tested, but its first
-// consumer (the `/photos/search/unified` endpoint) lands in Phase 2. Mirrors
-// llm_client.rs's allow-until-wired pattern so the bin target stays
-// clippy-clean in the interim; remove when the endpoint is added.
-#![allow(dead_code)]
-
 use crate::ai::llm_client::{ChatMessage, LlmClient, Tool, strip_think_blocks};
 use anyhow::{Result, anyhow};
 use serde::{Deserialize, Serialize};