ai: collapse llamacpp into LLM_BACKEND env switch

Reverts the per-request backend="llamacpp" value. Chat/vision/embedding
backend is now a deploy-time decision (LLM_BACKEND=ollama|llamacpp),
applied globally across chat, vision describe, and embeddings — so
embedding vectors stay in one space across the index.

- Per-request backend whitelist back to "local"|"hybrid". A request
  arriving with backend="llamacpp" is rejected.
- LLM_BACKEND=llamacpp swaps the entire local stack to llama-swap:
  chat hits the chat slot, describe hits the vision slot, embeddings
  hit the embed slot. Hybrid mode still routes chat to OpenRouter
  but uses LLM_BACKEND for the describe pass.
- Drops env vars HYBRID_VISION_BACKEND, LLAMA_SWAP_VISION_MODELS,
  EMBEDDING_BACKEND (the last never shipped). Drops the
  LlamaCppClient.vision_models allowlist — capability inference now
  reports has_vision only for the configured vision_model slot.
- Drops the /insights/llamacpp/models handler. /insights/models is
  the single endpoint; returns Ollama servers under LLM_BACKEND=ollama
  and llama-swap slots (from LLAMA_SWAP_ALLOWED_MODELS) under
  LLM_BACKEND=llamacpp. Same envelope shape either way.
- New ai::embed_one helper routes embeddings through llama-swap when
  LLM_BACKEND=llamacpp (else Ollama). Wires it into the four
  insight_generator embedding sites.
- Cross-replay matrix simplifies to pre-llamacpp shape (local↔local,
  hybrid↔hybrid, hybrid→local allowed; local→hybrid rejected).

.env.example

@@ -53,26 +53,30 @@ AGENTIC_CHAT_MAX_ITERATIONS=6
 # OPENROUTER_HTTP_REFERER=https://your-site.example
 # OPENROUTER_APP_TITLE=ImageApi
 
+# ── AI Insights — local backend switch ──────────────────────────────────
+# Picks which local LLM stack the server uses for chat, vision describe,
+# and embeddings. `ollama` (default) uses the OLLAMA_* settings above;
+# `llamacpp` uses the LLAMA_SWAP_* settings below. The switch is global
+# and applies to both `backend=local` and `backend=hybrid` (hybrid keeps
+# chat on OpenRouter but still uses this stack for the describe pass).
+# Don't flip mid-deploy without re-embedding existing index rows —
+# mixed vector spaces break similarity search.
+# LLM_BACKEND=ollama
+
 # ── AI Insights — llama.cpp / llama-swap (optional) ─────────────────────
-# Set LLAMA_SWAP_URL to enable the `llamacpp` chat_backend. Talks
-# OpenAI-compatible /v1 to a llama-swap proxy that fronts per-slot
-# llama-server instances (chat / vision / embed). Like hybrid, the
-# agentic loop describes images via the vision slot then inlines the
-# text into the chat slot — so the chat slot itself can be text-only.
+# Set LLAMA_SWAP_URL plus LLM_BACKEND=llamacpp to swap the local stack
+# off Ollama. Talks OpenAI-compatible /v1 to a llama-swap proxy fronting
+# per-slot llama-server instances (chat / vision / embed). The chat slot
+# is treated as text-only — images are pre-described via the vision slot
+# and inlined into the prompt.
 # LLAMA_SWAP_URL=http://localhost:9292/v1
 # LLAMA_SWAP_PRIMARY_MODEL=chat
 # LLAMA_SWAP_VISION_MODEL=vision
 # LLAMA_SWAP_EMBEDDING_MODEL=embed
-# Comma-separated allowlist of model ids the /v1/models endpoint should
-# advertise as vision-capable (llama-swap doesn't report modality).
-# LLAMA_SWAP_VISION_MODELS=vision
-# Comma-separated allowlist surfaced by /insights/llamacpp/models.
+# Comma-separated allowlist surfaced by /insights/models when
+# LLM_BACKEND=llamacpp.
 # LLAMA_SWAP_ALLOWED_MODELS=chat,vision,embed
-# LLAMA_SWAP_REQUEST_TIMEOUT_SECONDS=120
-# Routes hybrid mode's vision-describe pass through llama-swap's vision
-# slot instead of Ollama (chat still goes to OpenRouter). Values:
-# `ollama` (default) | `llamacpp`.
-# HYBRID_VISION_BACKEND=ollama
+# LLAMA_SWAP_REQUEST_TIMEOUT_SECONDS=180
 
 # ── AI Insights — sibling services (optional) ───────────────────────────
 # Apollo (places, face inference, CLIP encoders). Single-Apollo deploys