docs(L5): Reverse RAG framing + no-embeddings + per-1k COGS at $0.33/kWh

- Frame the solution as a Reverse RAG (Merkle Providence Reverse RAG) with
  a link to the whitepaper (unfirehose.com/merkle-providence-reverse-rag-
  whitepaper — note: published on unfirehose, not uncloseai).
- New differentiator: NO vector embeddings — retrieval is lexical-first
  (FTS5 BM25 + Merkle), dense-vector optional + off by default; embedding
  10M docs costs 10-100x more/doc + a vector index to store/maintain. A
  big part of why COGS is low.
- COGS framed per 1,000 answers, labeled @ $0.33/kWh (intro + diagram cost
  node). Cost node clarified: no embeddings, NO reasoning (reasoning is the
  thing that would cost 4-6x, which we skip). Retrieval node + mapping
  table updated to lexical-first / no vector index.

docs/_source/google-l5-rag-10m.rst

@@ -6,12 +6,16 @@ made the rounds online:
 
    *"Design a RAG pipeline for 10M docs with zero hallucination."*
 
-The canonical answer is a ten-box pipeline — ingest/normalize, hybrid
-BM25+embedding retrieval, ANN+rerank, source-confidence scoring,
-constrained generation, citation-backed responses, plus evals, caching and
-observability. arborist already implements every one of those boxes — and
-goes three steps further, which is what actually buys *zero hallucination*
-and a near-zero bill:
+arborist answers it as a **Reverse RAG** — the Merkle Providence Reverse
+RAG architecture (`whitepaper`_): instead of trusting a model and hoping
+it stays grounded, every claim is *bound back* to source spans by Merkle
+proof, and what cannot be bound is refused. The canonical answer is a
+ten-box pipeline — ingest/normalize, hybrid BM25+embedding retrieval,
+ANN+rerank, source-confidence scoring, constrained generation,
+citation-backed responses, plus evals, caching and observability.
+arborist already implements every one of those boxes — and goes four
+steps further, which is what actually buys *zero hallucination* and a
+near-zero bill:
 
 #. **A deterministic verifier, not a model confidence score.** "Zero
    hallucination" is not a threshold you tune — it is a property you prove.
@@ -21,10 +25,20 @@ and a near-zero bill:
 #. **A Merkle-bound cache that skips the GPU.** A hot answer is a
    content-addressed providence record with a Merkle proof — it replays
    with **zero GPU joules** and never re-enters the cost.
-#. **Measured energy COGS.** Constrained, non-reasoning generation
-   (cheap prefill + short decode, no test-time-compute chains) lands a
-   grounded answer for **~$0.07–0.16 per 1,000 answers** of GPU
-   electricity (see :doc:`bench`).
+#. **No vector embeddings.** Retrieval is **lexical-first** (FTS5 BM25 +
+   Merkle proofs); dense-vector semantic search is an *optional* layer,
+   **disabled by default**. Embedding 10M docs would cost **10–100× more
+   per document** to ingest and a whole vector index to store, maintain
+   and re-embed on drift — arborist skips all of it. That asymmetry is a
+   large part of why the bill is what it is.
+#. **Measured energy COGS — per 1,000 answers.** Because the pipeline is
+   a Reverse RAG with no embedding step and no reasoning chains (cheap
+   prefill + short decode), the GPU cost of a *grounded* answer is small
+   enough to **measure per thousand**: **~$0.07–0.16 per 1,000 answers
+   at $0.33/kWh** (see :doc:`bench`). A cache hit skips the GPU entirely,
+   so it never enters that count.
+
+.. _whitepaper: https://unfirehose.com/merkle-providence-reverse-rag-whitepaper
 
 Pipeline
 --------
@@ -71,7 +85,7 @@ Pipeline
      subgraph cluster_retr {
        label="3  RETRIEVAL + RERANK  (4 routes, merged)";
        style=dashed; color="#00aa00";
-       retr   [label="FTS5: BM25 body / title-LIKE /\ncore-keyword (TF-IDF) / phrase-pattern\n-> merge"];
+       retr   [label="FTS5 LEXICAL-FIRST (no vector index)\nBM25 body / title-LIKE /\ncore-keyword (TF-IDF) / phrase-pattern\n-> merge"];
        rerank [label="sqrt body-coverage rerank /\ntitle boost / rivalry exclusion +\nsynonym / per-source cap"];
        ctx    [label="context assembly\nwikitext -> base prose /\nper-mode budget"];
        retr -> rerank -> ctx;
@@ -82,7 +96,7 @@ Pipeline
        style=dashed; color="#aa00aa";
        gen  [label="claim_lattice (pointer IDs ->\nruntime-interpolated spans)\nOR quote - no test-time compute",
              style=filled, fillcolor="#fff0ff"];
-       cost [label="GPU: cheap prefill + short decode\n~$0.07-0.16 / 1k answers\n(reasoning chains cost 4-6x)",
+       cost [label="GPU: cheap prefill + short decode\n~$0.07-0.16 / 1k answers @ $0.33/kWh\nno embeddings · NO reasoning (would cost 4-6x)",
              shape=note, style=filled, fillcolor="#ffffcc"];
        gen -> cost [style=dotted, arrowhead=none];
      }
@@ -138,8 +152,9 @@ How arborist maps onto (and extends) the canonical design
      - **Merkle commitment** + idempotent content-addressing + lossless
        ``supersedes`` version history
    * - Hybrid retrieval (BM25 + embeddings)
-     - 4-route FTS5 merge (``qa/query.py``)
-     - phrase-pattern route (closes the allusion gap) + rivalry exclusion
+     - 4-route FTS5 merge, **lexical-first** (``qa/query.py``)
+     - **no embeddings needed** — dense-vector optional + off by default
+       (10-100x cheaper ingest, no vector index); phrase-pattern route
    * - ANN + reranking
      - sqrt body-coverage rerank + title boost
      - corpus-derived synonym/rivalry layer, no embedding index to drift