fix: remove node_modules from /infrabox/cache to prevent timeout

The real bottleneck was not cp -r in build.sh but job.py's post-build
step that compresses and uploads /infrabox/cache via snappy+tar to the
API server. With node:20's much larger node_modules (~200MB+), this
compression/upload exceeds the 1-hour job timeout.

Solution: stop writing node_modules back to /infrabox/cache. Instead,
use mv to restore cached node_modules at the start (fast), and don't
write it back (npm install with warm cache only takes ~20s anyway).
This eliminates the expensive cache upload entirely.

Author: Yuan Huang

src/dashboard-client/build.sh

@@ -2,8 +2,11 @@
 cp -r /infrabox/context/src/dashboard-client /dashboard
 
 echo "## Link cache"
-mkdir -p /infrabox/cache/node_modules
-tar -C /infrabox/cache -cf - node_modules | tar -C /dashboard -xf -
+if [ -d /infrabox/cache/node_modules ]; then
+    mv /infrabox/cache/node_modules /dashboard/node_modules
+else
+    mkdir -p /dashboard/node_modules
+fi
 
 cd /dashboard
 
@@ -15,10 +18,6 @@ npm install
 echo "## build"
 npm run build
 
-echo "## Copy to cache"
-rm -rf /infrabox/cache/node_modules
-tar -C /dashboard -cf - node_modules | tar -C /infrabox/cache -xf -
-
 echo "## Copy to output"
 cp -r /dashboard/dist /infrabox/output