feat(embed): candle Metal (Apple Silicon GPU) opt-in build feature + v0.23.0

- kebab-embed-candle: `metal` feature → candle metal backend; select_device()
  picks Device::new_metal(0) (CPU fallback) under the feature, else Device::Cpu.
  .contiguous() before to_vec2 (Metal rejects strided views; CPU tolerates).
- feature passthrough: kebab-app/embed_metal → kebab-cli/embed_metal.
  Build on macOS: cargo build --release --features embed_metal.
- default (non-metal) path unchanged: clippy 0, candle units + thread_cap + parity pass.
- README + HOTFIXES: Mac-GPU-ingest → copy sqlite+lancedb → server CPU-query workflow.
- version 0.22.0 → 0.23.0 (opt-in build surface).

macOS-only compile; Metal execution/speed/parity validated by user on M4 Pro
(not buildable on the Linux CI/dev machine).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-06-02 11:37:08 +00:00
parent 99f8cfa691
commit 369aeb3d24
8 changed files with 485 additions and 50 deletions

View File

@@ -31,7 +31,7 @@ edition = "2024"
rust-version = "1.85"
license = "MIT OR Apache-2.0"
repository = "https://github.com/altair823/kebab"
version = "0.22.0" # v0.22.0 — candle 임베딩 provider (NUMA-안전, opt-in `provider=candle` + `num_threads`/KEBAB_EMBED_THREADS). fastembed default 불변, embedding_version 유지(재색인 0). — CLAUDE.md §Release 도그푸딩 트리거
version = "0.23.0" # v0.23.0 — candle Metal(Apple Silicon GPU) opt-in build feature (`--features embed_metal`): M-series 맥에서 e5-large 임베딩을 GPU 로 → 대용량 ingest 가속. macOS 전용, 벡터는 CPU candle 과 호환. default(비-metal) 동작 불변. — CLAUDE.md §Release 도그푸딩 트리거
# pre-v0.18 workspace-wide cleanup: enable clippy::pedantic group with
# intentional allow-list. The allowed lints are either cosmetic (doc style),