Fix handler of forgejo restart running when service doesn't exist yet

roles/forgejo/handlers/main.yml

@@ -22,17 +22,6 @@
   changed_when: false
   failed_when: '"Good signature from" not in result.stderr'
 
-- name: Copy binary to global location
-  become: true
-  ansible.builtin.copy:
-    src: "{{ forgejo_run_dir }}/forgejo-{{ forgejo_latest_version }}-linux-{{ forgejo_architecture }}"
-    dest: /usr/local/bin/forgejo
-    remote_src: true
-    owner: root
-    group: root
-    mode: 0755
-  notify: Restart forgejo service
-
 - name: Start forgejo service
   become: true
   ansible.builtin.systemd:

roles/forgejo/tasks/binary.yml

@@ -32,7 +32,18 @@
     - Receive forgejo pgp key
     - Download forgejo asc file
     - Verify forgejo binary with gpg
-    - Copy binary to global location
 
-- name: Make sure downloaded binary has been verified and copied to global location
+- name: Make sure downloaded binary has been verified
   ansible.builtin.meta: flush_handlers
+
+- name: Copy binary to global location
+  become: true
+  ansible.builtin.copy:
+    src: "{{ forgejo_run_dir }}/forgejo-{{ forgejo_latest_version }}-linux-{{ forgejo_architecture }}"
+    dest: /usr/local/bin/forgejo
+    remote_src: true
+    owner: root
+    group: root
+    mode: 0755
+  when: forgejo_installed_version.stdout is version(forgejo_latest_version, '<')
+  notify: Restart forgejo service