laya: reuse native input staging and group CUDA uploads

[Levius-Fubuki]

Purpose

Reuse workspace host staging for IDs, lengths and question types, then submit one validated group with one synchronization. Foreign buffers or an oversized member reject the entire group before CUDA. Attempted submissions are drained even after copy failure while host borrows remain live; empty groups enter no native copy/sync functions. Buffer::write uses the same drain path.

Stack 2/4, dependent on #71; merge in order. This PR targets main to run current-head CI and includes predecessor increments until they merge. Review this increment against 5833efab791f9f53669690c96e6ad90fa6b2f5a3.

Test Plan

Published head: 3923c1e0f16f74f31252b9dc7e2304c81284ee0d
Validated source revision: 6ecf4a11bff5a71e81215a980b4389a88c54600e (source-identical tree; the published head adds only an empty CI-trigger commit)
Incremental baseline: 5833efab791f9f53669690c96e6ad90fa6b2f5a3
Current-main baseline: 7f39ac40902c374803992407bb26eeba29c8a588

cargo fmt --all --check
cargo clippy --workspace --locked --all-targets --offline --features omni-laya/serve -- -D warnings
cargo test --workspace --locked --offline --features omni-laya/serve
cargo build --workspace --release --locked --offline --features omni-laya/serve

Review incremental and cumulative diffs. Existing history is preserved: the previous PR head remains an ancestor. Current full-request processing/heads/HTTP/runtime integration replaces obsolete encoder-only modules.

Test Result

All four checks passed on the exact individual validated source revision above. The published head has the same tree, confirmed with Git; its current-head CI is checked separately. 83 passed, 0 failed, 10 ignored. Ignored checkpoint/GPU tests remain unverified. Incremental git diff --check passed. The PR remains core source only; external fixtures/logs/evidence are retained locally.

External model fixtures passed unchanged launch traces, staging reuse, one input-group sync, and A→B→A IDs/lengths/types updates. CPU ABI fixtures passed foreign/bounds prevalidation, empty groups, copy-failure drain and combined copy/sync errors; real-driver grouped-copy checks passed.

Fresh generic CUDA validation passed using exact #73 backend source 231cf358770e1464efbbce86ec6ce47cb5746056 on RTX 4090 (sm89), driver 580.105.08, nvcc 13.0.88 and Rust 1.99.0. An external init-only sm89 bootstrap and scalar-add kernel exercised production resource/copy/capture/replay/free operations. The production Hopper gate was preserved. #71 resolved dispatch matches tested methods byte-for-byte; #72 group copy matches outside the later capture guard; #74 backend source equals #73. Two setup attempts stopped before GPU work because Python was missing from PATH; corrected PATH passed with unchanged source/assertions.

Validation limit: This is generic host/backend FFI evidence, not Laya support on RTX 4090. Full optimized Laya eager/Graph/scorer/action-head/HTTP numerical parity and performance were not rerun on Hopper after this port. No new Laya accuracy or speedup claim is made. The earlier 2026-10-04 H800 encoder-stack comparison remains historical.

Numerical CUDA/TileLang/RoPE implementation, operator export/build sources, weights/conversion, preprocessing, padding, decoding, executor/serving and Cargo metadata match current main. Only host execution/resources/cache and scoped C capture cleanup change.

Self-review

Incremental and cumulative diffs were reviewed against current architecture/runtime contracts, supported inputs, precision, pointer/drop lifetimes, failure paths and claims. Local review found no remaining actionable defect in this host scope. Fresh full-Hopper validation remains unverified; contributor and maintainer review are separate from this assistance.

Demo / evidence

Raw current-source workspace, CPU trace/copy/capture/cache fixture logs, exact source/ancestry hashes and real-driver output are retained under artifacts/all-pr-followups-20261005/laya/, outside this core-code diff. The deliberately caught panic in the driver log is an expected cleanup test; the suite exit code is zero. No video or measured speedup is claimed. Current-head GitHub Rust, benchmark and docs CI all passed; deploy skipped. CI run.

[Copilot]

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