Rebuild lean-bench: pages in, verified indexed Lean out

Replaces the previous tree entirely. The pipeline is four separable pieces, each
testable on its own:

1. leanbench/llm.py      provider-neutral OpenAI-compatible connector, stdlib only,
                        with retries, JSON coercion and a scripted client so the
                        rest of the code is testable without a network.
2. leanbench/detect.py   per-page detection of formalizable mathematics. Case 1 is
                        mathematics asserted as fact (definition/theorem/lemma/...,
                        deliberately not sub-classified); case 2 is a problem left
                        open to the reader. Hits come back as spans -- start page
                        and index, end page and index -- so several per page is
                        normal. A hit cut off by the page end makes the detector
                        re-read that page together with the next, repeatedly, so an
                        item spread over three pages is found whole.
                        Offsets come from verbatim quotes resolved by
                        leanbench/anchors.py (exact -> whitespace-insensitive ->
                        fuzzy), because models quote well and count badly.
3. leanbench/formalize.py + leanbench/verify.py
                        each span becomes one Lean 4 declaration, judged by the real
                        compiler: full elaboration for case 1, statement legality
                        only for case 2, which has no proof to check. Lean's own
                        diagnostics feed back for a repair attempt. "verified" (Lean
                        accepted it) and "proved" (accepted with no sorry) are kept
                        strictly apart.
4. leanbench/db.py       indexed SQLite plus JSONL export. Every item is indexed by
                        the types of its inputs, or by its own type when it is not a
                        proof, and records the source book (sha256) and the page it
                        starts on -- so `drop --source H --pages A-B` followed by a
                        re-ingest redoes exactly those pages and nothing else.

Also: lean/ Lake project pinned to Mathlib, docker/ image carrying Lean 4 and a
prebuilt Mathlib (the Mathlib revision and the toolchain are resolved against each
other at build time rather than both hardcoded), and .gitea/workflows for tests,
image build and push to the Gitea registry, and a manual "submit a book" run.

58 tests; they need neither network nor Lean.

lean/LeanBench.lean

@@ -0,0 +1,18 @@
+/-
+The verification environment.
+
+Candidate declarations are never added to this file: each one is elaborated in its
+own scratch file against this project's imports, so a broken candidate can never
+break the environment the next candidate is checked in.
+
+Anything you want every candidate to see -- a shared convention, a local notation --
+belongs here.
+-/
+import Mathlib
+
+namespace LeanBench
+
+/-- Sanity check: if this fails, the toolchain and Mathlib do not agree. -/
+theorem environment_ok : (2 : ℕ) + 2 = 4 := by norm_num
+
+end LeanBench

lean/lakefile.toml

@@ -0,0 +1,12 @@
+name = "leanbench"
+defaultTargets = ["LeanBench"]
+
+# The revision is rewritten by docker/resolve-mathlib.sh at image build time, so the
+# Mathlib revision and the Lean toolchain always agree. Edit MATHLIB_REV, not this.
+[[require]]
+name = "mathlib"
+git = "https://github.com/leanprover-community/mathlib4.git"
+rev = "v4.33.1"
+
+[[lean_lib]]
+name = "LeanBench"

lean/lean-toolchain

@@ -0,0 +1 @@
+leanprover/lean4:v4.33.1