Github Actions update (#2951)

Update setup-node action, release-drafter action and remove manual docker-compose step to fix tests.

Author: amCap1712

.github/workflows/frontend-tests.yml

@@ -38,7 +38,7 @@ jobs:
     # We do not use this to install node but only to register problem matchers
     # so that eslint annotations work.
     - name: Setup Node.js environment to enable annotations
-      uses: actions/setup-node@v3
+      uses: actions/setup-node@v4
 
     - name: Build frontend containers
       run: ./test.sh fe -b

.github/workflows/release-drafter.yml

@@ -11,6 +11,6 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       # Drafts your next Release notes as Pull Requests are merged into "master"
-      - uses: release-drafter/release-drafter@v5
+      - uses: release-drafter/release-drafter@v6
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/unit-tests.yml

@@ -43,11 +43,5 @@ jobs:
       run: echo ${{ secrets.DOCKER_HUB_PASSWORD }} | docker login -u ${{ secrets.DOCKER_HUB_USERNAME }} --password-stdin
       continue-on-error: true
 
-    # by default, docker-compose tries to pull the `listenbrainz` image used by timescale_writer as well.
-    # but this is built locally for the listenbrainz service and reused for timescale writer, hence the pull fails.
-    # to avoid this we specifically mention the services to pull images for here,
-    - name: Pull docker images
-      run: docker-compose -f docker/docker-compose.test.yml pull lb_db redis rabbitmq couchdb
-
     - name: Run tests
       run: ./test.sh