Avoid hardcoding remote names in the sync jobs (#33)

Use new version of https://github.com/paradedb/actions to avoid
hardcoding remote names.

Author: stuhood

.github/workflows/sync-promote-branch.yml

@@ -19,7 +19,7 @@ concurrency:
 
 jobs:
   call-promote:
-    uses: paradedb/actions/.github/workflows/reusable-promote.yml@v4
+    uses: paradedb/actions/.github/workflows/reusable-promote.yml@v5
     with:
       branch_name: ${{ inputs.branch_name }}
       approvers: "philippemnoel,rebasedming,stuhood,mdashti"

.github/workflows/sync-upstream-rebase.yml

@@ -17,7 +17,7 @@ concurrency:
 
 jobs:
   call-rebase:
-    uses: paradedb/actions/.github/workflows/reusable-rebase.yml@v4
+    uses: paradedb/actions/.github/workflows/reusable-rebase.yml@v5
     with:
       github_app_client_id: ${{ vars.PARADEDB_GITHUB_APP_CLIENT_ID }}
     secrets:

scripts/sync-upstream.sh

@@ -14,7 +14,7 @@ export UPSTREAM_BRANCH="main"
 
 # 2. Define the URL to the centralized script
 # Using the raw content URL from the central repository
-CORE_SCRIPT_URL="https://raw.githubusercontent.com/paradedb/actions/v4/scripts/sync-core.sh"
+CORE_SCRIPT_URL="https://raw.githubusercontent.com/paradedb/actions/v5/scripts/sync-core.sh"
 
 # 3. Download and source the core logic as an API
 TMP_SCRIPT=$(mktemp)