Run reviewdog directly

So that we can control the version we run without depending on external
actions, and so that we use a consistent version for all linters.

At the same time, unify to running the latest version of reviewdog,
0.21.0, which also involves changing the deprecated `fail_on_error` flag
previously used by prettier action to `fail_level`.

.github/workflows/linters.yml

@@ -22,21 +22,10 @@ jobs:
 
       - run: git show --no-patch # the commit being tested (which is often a merge due to actions/checkout@v3)
 
-      - name: prettier
-        uses: EPMatt/reviewdog-action-prettier@v1
+      - uses: reviewdog/action-setup@v1
         with:
-          github_token: ${{ secrets.github_token }}
-          reporter: github-pr-check
-          level: error
-          filter_mode: nofilter
-          fail_on_error: true
+          reviewdog_version: v0.21.0
 
-      - name: rubocop
-        uses: reviewdog/action-rubocop@v2
-        with:
-          skip_install: true
-          reporter: github-pr-check
-          level: error
-          filter_mode: nofilter
-          use_bundler: true
-          fail_level: any
+      - run: ./script/reviewdog.sh
+        env:
+          REVIEWDOG_GITHUB_API_TOKEN: ${{ secrets.github_token }}