spec(port): port the commit secret export collision guard (Refs #8883)

[dmitrii-f-t27]

The second CR-CHAT-03 ring target (W3, tracker #8883, epic gHashTag/trios#1812): port of crates/trios-chat/rings/CR-CHAT-03/src/commit_secret_export_collision.rs (Wave-33, L-CHAT-3-csec, R-CHAT-11, RFC 9420 8.4/9).

After each Commit the deriving member exports a fresh commit secret from the KDF tree, bound to the group, the epoch and the commit transcript hash. A hostile member splices across groups or epochs, replays a published secret, binds to a phantom transcript or submits the zero sentinel; a single deny wins. Rules, in the original's fixed order:

1. canonical secret length (32),
2. non-empty transcript hash,
3. transcript present in the local ledger,
4. group_id == expected_group_id,
5. commit_epoch == current_epoch,
6. (transcript_hash, commit_secret) pair fresh against every prior export,
7. commit_secret not the all-zero sentinel.

Carried as the D5 verdict — all seven original error variants carry no payload, so the code is the whole verdict.

Adaptations (t27 conventions): export/view structs → flat parameters, byte vectors as plain slices; the vector of known transcript hashes → slice of fixed-capacity records (64 bytes, the original's generous upper bound); the vector of exported pairs → slice of records holding transcript and secret with lengths; the original's Vec/tuple contains → scans. The known-hash record's length field is named hash_len — a field named len anywhere in the spec breaks gen-rust's slice-length lowering (the DMPE lesson, bitten again here).

All 10 original CSEC tests transcribed with exact asserts (CSEC-10 pins that the replay rule reads the pair, never the hash alone), plus 4 helper tests and a rule-order test stacking each earlier rule's violation on a later one (15 total).

Checks (conventions 0-14): t27b test 15 passed 0 failed, 0 blockers; t27c test-report 15/15, 0 vacuous; gen-rust output compiles under rustc (0 'static str, 0 Vec<const); coverage Untested 0; ASCII only; seal saved, all hashes MATCH; ledger row added to docs/reports/t27b_expectations.json.

Refs #8883

[dmitrii-f-t27]

All required checks green (validate, check-linked-issue, parse-ratchet, t27b-native-linux, t27b-native-macos); cli-tri / duplicate-bodies / t27b-native-ratchet are the known non-required reds. Steward READY. Second CR-CHAT-03 target (2/47 T-PORT) — requesting owner merge when convenient, Refs #8883.

[dmitrii-f-t27]

Superseded by #10175: the remaining CR-CHAT targets now land as one port-wave batch per specs/policy/pr_budget.t27 (#10067/#10074), which asks for one PR per ring batch instead of one per file. The spec body, seal and ledger row in this PR are unchanged there; the batch also carries one self-consistent duplicate-body bless. Closing this one-file PR keeps the merge queue and the Actions queue small.