feat(decider): independent CUDA Graph replay mode

[Levius-Fubuki]

Purpose

Add Decider's strict default-off DECIDER_GRAPH=0|1 switch, independent of CUA_S1_GRAPH. An explicit shared Qwen loader preserves the original constructor behavior for existing workers. Cache backbone captures by ordered row-length vectors (max64 FIFO); upload current IDs before replay and synchronize before scratch growth or disposal. Head projection and calibrated responses remain outside the graph. Capture failure falls back to eager for the worker lifetime.

Actual capture/replay/fallback/invalidation counters remain available in health through a separate statistics snapshot. Decision responses are unchanged. Depends on #111, #110 and #94; the main-based diff includes their core implementations.

Submission scope

Core implementation and directly related tests are retained. Ordinary documentation, experiment reports/media and generated run outputs remain outside this submission. Test fixtures, build inputs, deployment data and third-party licensing remain as needed. The diagnostic test helper lives at tests/decider/runner.rs, built with cargo build -p omni-decider-native --example decider-run --release --locked; its executable is target/release/examples/decider-run. It is not a production worker binary.

Validation

After cleaning workspace debug artifacts before each branch, fresh local fmt, Clippy/all-targets with warnings denied and 110 Rust tests pass (22 opt-in tests ignored). Independent GitHub CI passes on this exact head, including release workspace build and strict Docs. Restored test bodies/fixtures match the saved originals. Production execution is unchanged. GPU tests were not rerun; the server remains shut down. 2 Python regression tests also pass.

Head 25581a563d936a65dd810542f3f8d95b9cee7d6a; 42 changed files and 3,122 changed source/test/build lines, excluding licenses, notices, locks and static data.

Self-review

Full focused diff and lifetime/caller contracts reviewed locally and independently; corrected both health integration findings before publishing. Generated run output remains outside committed source. Requested full agent-assisted self-review is complete; maintainer review remains separate.

Main-based diff:3118authored lines including manifests,3846total. This exceeds3000due to pending parent contracts/worker/batching. Full component self-review covered CPU contract/checkpoint/processor and serial worker/head in #94/#110, packing/retirement in #111, and explicit Graph/lifetime/health snapshot integration here. Split into dependent focused drafts; review those parents first, then the d88179c Graph commit. All affected normal workspace/HTTP/device boundaries validated as described; no unreviewed generated outputs committed.

Final followup evidence

Immutable sequential validation archive, asset decider-next-20261007-final-evidence.tar.gz; SHA256 9f57d969fc484c518d8ebac7b2dd38174e2c6954e7af04b9352550675053ae48. Includes exact5branch heads, final source-vs-executed-runtime audit, frozen protocols, raw repeated samples/responses, failures, source/binary/library hashes, environment and full self-review. Weights/credentials/private inputs excluded.

Final heads: contract f59cca40462d5b63c3dbc909e6aefb147bd8f44e; worker 9a09e5a58d33692edd5aab5634f268055e4c8151; batching 19b262e544de2c9e2d0b23f865204a6ae2372038; Graph d88179c6f3b78cd33025031134b7a8086e61f481; dependent prefix 6962971820002d46b030e09503cd21d44c831083. Main base 99865743d27316fbe81362dc0f0e6de6fda86284. All current-head CI Rust/benchmark/docs checks passed; docs deploy intentionally skipped on PRs. Runtime source/manifests match measured snapshots byte-for-byte; final prefix head adds a documentation-only command repair.

2026-10-08 verification follow-up

Validated source head: e13b76748fcce7f4881289709af4f31e964cbf65. Upstream main snapshot: 4a79980d; GitHub reports no merge conflict. Fresh exact-head fmt, strict workspace Clippy, workspace tests and release build pass: 110 passed, 0 failed, 22 ignored. Ignored tests are counted separately from explicit GPU execution. Exact commands and logs are in the archive. Merged current main Open-Jev-9B support and resolved documentation/index conflicts while retaining both model entries. Strict MkDocs also passes. GPU/reference results above were executed on snapshot d88179c6f3b78cd33025031134b7a8086e61f481; a complete source comparison confirms all affected model/runtime/backend and regression sources are byte-identical at this new integration head (see gpu-source-binding.json). Open-Jev-9B inference is not newly validated by these GPU cases. Fresh GPU execution passes the 3 Decider integration regressions, 2 BF16 head tests and both shared-Qwen Graph regressions. Current-input replay, ordered shape keys, growth/FIFO and legacy constructor behavior are exercised. Keep dependent on #111/#110/#94; prior repeated mode/HTTP/fault evidence remains in its original dated archive.

Raw verification evidence, SHA-256 5df66bf3974d534875946168fba4d7fe543d08edbd658c3756f17d6f0c8f5773. Includes exact source manifests, environment/library/binary hashes, complete logs, synthetic reference/HTTP results and initial setup failures. CUDA execution is serial on one RTX 4090. No weights, credentials or private inputs are included. Contributor/maintainer review and dependency merges remain separate.

Completed self-review checklist

• I have reviewed the full diff and addressed the issues I found.
• I have checked that the change follows the project's architecture and stays focused on the stated purpose.
• I have run the checks appropriate to this change and reported commands, results, and anything I could not verify above.
• I have checked that the PR description, documentation, and any accuracy or performance claims match the implementation and available evidence.

Full self-review completed / Ready — 2026-10-08

Current head: b9a0051761d11aaf4005b53703dbc163199c9dd3; merge base: 4a79980d8a75fd063cb3f8247e06215e288b18ac. Full cumulative diff: 3118 authored-code lines / 3846 total diff lines. The author explicitly requested completed self-review and Ready conversion. The completed checklist records this agent-assisted review and its verification; it does not represent maintainer approval.

Root integration review plus three independent component reviews covered all changed source, tests, helpers, manifests and documentation: CPU contract/checkpoint/tokenizer/calibration; serial worker/admission/GPU head and error retirement; ordered bounded packing and response reconstruction; Graph lifetime/cache/health/options; ABI7 CUDA continuations/fixed GEMM/request-local prefix state; root regressions, recipe reproduction, fixture consumers and evidence/claim boundaries. No functional blocker remains. For #112/#113 above 3,000 authored lines, this is full component/integration self-review, with the existing split rationale and review order retained.

Corrected stale unsupported-Graph and no-worker-prefix-consumer documentation where applicable. Removed unused CPU fixture run/duplicate fields on the Decider stack, preserving all 255 label IDs and all 12 used fixture values exactly; the original removed data remain in evidence outside committed source. Both pinned CPU contract opt-ins pass on every repaired Decider branch; repaired docs pass strict MkDocs. All five initial review heads passed fresh fmt, strict workspace Clippy, workspace tests, release build and strict docs; latest-head CI Rust/benchmark/docs also passes. Ignored hardware/model tests are separate from executed tests.

Runtime/CUDA/GPU regression code is unchanged by this cleanup. Earlier personally generated GPU/reference/HTTP campaigns retain their original source revisions; independent reviewers rehashed both source bindings, replayed all stored numerical gates and fixed/shared exact controls, and recomputed published timings. ready-source-binding.json binds these runtime checks to the latest heads and separately proves the used CPU oracle values are unchanged. No new CUDA/HTTP/performance campaign was run after server shutdown; the server remains off. Existing hardware, scope and performance limitations remain disclosed.

Complete self-review reports and fresh checks; SHA-256 1edbd4a6e13576d4cc23638b7ec1e35c1248d6a712b8a6406e1f269ca685cb9d. Includes all four review reports, closure summary, original reviewed full diffs, new CPU/docs logs, latest-head CI, source/fixture binding and archived removed fields. Merge dependencies remain #103 before #106, and #94 → #110 → #111 → #112 → #113, including #97/#98/#99 prerequisites for #113; Ready conversion does not merge them.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.