Add --http-only and --browser-only scheduler flags for split deployment

Allow running separate crawler instances for HTTP and browser work:
- --http-only: skips browser phases, sets max_browser=0 (for lightweight fly.io workers)
- --browser-only: skips HTTP phases, sets max_concurrent=0 (for dedicated Hetzner browser worker)

Also updates fly.toml to use --http-only and reduces VM to 512MB,
adds Hetzner deploy workflow (GHCR + SSH), and .dockerignore.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: viktor-shcherb

.github/workflows/deploy-crawler-browser.yml

@@ -22,9 +22,6 @@ jobs:
         id: version
         run: echo "version=$(cat apps/crawler/VERSION | tr -d '[:space:]')" >> "$GITHUB_OUTPUT"
 
-      - name: Set up QEMU (for ARM64 cross-build)
-        uses: docker/setup-qemu-action@v3
-
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
 
@@ -39,7 +36,7 @@ jobs:
         uses: docker/build-push-action@v6
         with:
           context: apps/crawler
-          platforms: linux/arm64
+          platforms: linux/amd64
           push: true
           tags: |
             ghcr.io/${{ github.repository_owner }}/jobseek-crawler:latest