gpulock: GPU arbitration proxy for Ollama + ComfyUI

Implements SPEC.md: two listeners, one writer-preferring two-mode lock,
Ollama unload before image jobs, ComfyUI history polling + VRAM free,
optional model warm-up, healthz/metrics endpoints, streaming-safe
reverse proxies, Dockerfile and Gitea Actions CI.

Dockerfile

@@ -0,0 +1,13 @@
+FROM golang:1.23 AS build
+WORKDIR /src
+COPY go.mod ./
+COPY cmd ./cmd
+COPY internal ./internal
+ARG VERSION=dev
+RUN CGO_ENABLED=0 go build -ldflags="-s -w -X main.version=${VERSION}" -o /gpulock ./cmd/gpulock
+
+FROM gcr.io/distroless/static:nonroot
+COPY --from=build /gpulock /gpulock
+EXPOSE 8189 11435
+USER nonroot
+ENTRYPOINT ["/gpulock"]