Actually apply the pre-commit fixers to the codebase.

This can be redone manually with
`pre-commit run --all`

While the pre-commit hook could be merged to run locally,
it is much cleaner to align all the files to best-practice
syntax in a single commit. It is also required for server-side
validation.

.github/workflows/docs.yaml

@@ -7,11 +7,9 @@ jobs:
   publish-docs:
     runs-on: ubuntu-latest
     steps:
-      -
-        name: Checkout ntfy code
+      - name: Checkout ntfy code
         uses: actions/checkout@v3
-      -
-        name: Checkout docs pages code
+      - name: Checkout docs pages code
         uses: actions/checkout@v3
         with:
           repository: binwiederhier/ntfy-docs.github.io
@@ -19,18 +17,15 @@ jobs:
           token: ${{secrets.NTFY_DOCS_PUSH_TOKEN}}
           # Expires after 1 year, re-generate via
           # User -> Settings -> Developer options -> Personal Access Tokens -> Fine Grained Token
-      -
-        name: Build docs
+      - name: Build docs
         run: make docs
-      -
-        name: Copy generated docs
+      - name: Copy generated docs
         run: rsync -av --exclude CNAME --delete server/docs/ build/ntfy-docs.github.io/docs/
-      -
-        name: Publish docs
+      - name: Publish docs
         run: |
           cd build/ntfy-docs.github.io
           git config user.name "GitHub Actions Bot"
-          git config user.email "<>"          
+          git config user.email "<>"
           git add docs/
           git commit -m "Updated docs"
           git push origin main