Initial commit: Naut-Torrent — from-scratch 10 GbE BitTorrent client

A maintainable, extensible BitTorrent client (C11, Linux/io_uring) targeting
10 GbE saturation. All torrent functionality is built from scratch; liburing
is the only linked third-party dependency on the data path.

Implements Phases 1-7 of the roadmap:
- core: page-aligned buffer pool, MPMC/Treiber queues, bitfields, worker pool
- crypto: SHA-1/256 (SHA-NI + scalar), Merkle (BEP-52), RC4 (MSE)
- bencode/metainfo: zero-copy parser, v1/v2/hybrid .torrent + magnet
- peer: sans-IO wire codec, MSE/PE handshake state machine, BEP-10, ut_metadata, PEX
- piece/storage: block-level multi-peer engine, rarest-first + endgame,
  per-file completion events + single-file relocate (move-as-you-finish)
- tracker/dht: HTTP + UDP (BEP-15) trackers, BEP-5 KRPC iterative lookup
- platform: io_uring reactor (SQPOLL, registered buffers, SEND_ZC)
- surface: versioned RPC, native plugin ABI, sandboxed Lua scripting, nautd/nautctl

Verified against libtorrent (single/multi/hybrid, MSE, magnet-via-DHT, swarm);
unit + interop tests green; ASan/UBSan/TSan clean. Scripting reference in
docs/scripting.md.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

include/naut/storage.h

@@ -0,0 +1,46 @@
+/* storage.h — file backend mapping the torrent's flat byte space to files.
+ *
+ * A torrent is one contiguous byte space [0, total_length); this layer splits a
+ * read/write at any global offset across the underlying files (a single block
+ * write can straddle a file boundary). Phase 3 uses positional pread/pwrite for
+ * correctness; the io_uring O_DIRECT fast path is a Phase 6 swap behind this
+ * same interface. The backend is a vtable so a memory/object-store backend can
+ * be registered later (the "extensible" goal).
+ */
+#ifndef NAUT_STORAGE_H
+#define NAUT_STORAGE_H
+
+#include "naut/common.h"
+#include "naut/metainfo.h"
+
+typedef struct naut_storage naut_storage;
+
+typedef struct {
+    bool direct_io;
+    bool preallocate;
+} naut_storage_opts;
+
+/* Open (creating + preallocating) all files under `root`. */
+naut_storage *naut_storage_open(const naut_file *files, size_t nfiles,
+                                const char *root, naut_err *err);
+naut_storage *naut_storage_open_opts(const naut_file *files, size_t nfiles,
+                                     const char *root,
+                                     const naut_storage_opts *opts,
+                                     naut_err *err);
+void naut_storage_close(naut_storage *s);
+
+naut_err naut_storage_write(naut_storage *s, int64_t offset, const void *buf, size_t len);
+naut_err naut_storage_read(naut_storage *s, int64_t offset, void *buf, size_t len);
+naut_err naut_storage_sync(naut_storage *s);
+
+/* Move one completed file out to `dest` (rename, or copy+unlink across file
+ * systems). The caller must guarantee the file is complete — every piece
+ * overlapping it verified — so no further writes target it. After this the slot
+ * is "externalized": subsequent I/O to its region returns NAUT_ERR_RANGE. This
+ * is the storage half of the "move files as they finish" feature. */
+naut_err naut_storage_relocate(naut_storage *s, size_t file_index, const char *dest);
+
+int64_t naut_storage_total(const naut_storage *s);
+bool    naut_storage_direct_enabled(const naut_storage *s);
+
+#endif /* NAUT_STORAGE_H */