[Shellghostt]

Closes #4

What

Adds a production-lean Docker image for WebVac's scrape/crawl mode:

• Dockerfile — python:3.12-slim, installs the package from source (pip install .) and Patchright Chromium with its OS deps, runs as webvac with /data as the working directory so scraped_data/ can be volume-mounted.
• .dockerignore — keeps VCS metadata, virtualenvs, caches, tests, docs, the optional decaffeinator submodule, and local secrets/credentials out of the build context.

The optional De-Caffeinator VAPT task is intentionally excluded: it needs the submodule plus Node.js, which would bloat the image for the common scrape use case.

Usage

docker build -t webvac .
docker run --rm webvac --doctor
docker run --rm -v webvac-data:/data webvac --url https://example.com --mode single

Verification

Build:

$ docker build -t webvac:test .
...
#14 naming to docker.io/library/webvac:test done
build exit=0

Doctor preflight (real headless Chromium launch inside the container):

$ docker run --rm webvac:test --doctor
[OK] Python 3.12.16
[OK] Output directory writable: scraped_data
[OK] Patchright browser launched successfully.
Doctor summary: 3 ok, 4 warnings, 0 failures
doctor exit=0

CLI entrypoint:

$ docker run --rm webvac:test --help
usage: webvac [-h] [--url URL] [--doctor] ...
help exit=0