fix(deploy): switch from nixpacks to Dockerfile + change branch to main

Nixpacks auto-detect could not find a 'start' script in package.json
and bailed out. Astro builds to static files in dist/ — there is no
Node server to start. Switching to a Dockerfile + nginx fixes the
'No start command could be found' error from EasyPanel.

The workflow also pointed at the source-code branch, but the panel's
git source ref for the dealplustech-astro service is 'main', so the
trigger was firing for the wrong ref. Both workflows now run on push
to main.

- Dockerfile: multi-stage node:20-alpine build + nginx:1.27-alpine serve
- nginx.conf: gzip, security headers, 1-year cache for hashed assets,
  try_files fallback for UTF-8 slugs (Astro file-based routing)
- .dockerignore: keep build context small (skip CI, docs, .gitea, IDE)
- build-and-deploy.yml + lint.yml: branch source-code -> main
- docs/ci-setup.md: corrected project + service names (customerwebsite
  / dealplustech-astro), documented the Dockerfile rationale, added a
  note for the 'Failed to sync changes' server-side error

Dockerfile

@@ -0,0 +1,29 @@
+# =====================================================================
+# Stage 1: Build the Astro static site
+# =====================================================================
+FROM node:20-alpine AS builder
+
+WORKDIR /app
+
+# Install deps with cache layer
+COPY package*.json ./
+RUN npm ci --no-audit --no-fund
+
+# Build
+COPY . .
+RUN npm run build
+
+# =====================================================================
+# Stage 2: Serve with nginx
+# =====================================================================
+FROM nginx:1.27-alpine
+
+# Astro outputs to ./dist by default
+COPY --from=builder /app/dist /usr/share/nginx/html
+
+# nginx config: SPA-friendly, gzip, cache headers for static assets
+COPY nginx.conf /etc/nginx/conf.d/default.conf
+
+EXPOSE 80
+
+CMD ["nginx", "-g", "daemon off;"]