fix: revert autolabeler to release-drafter v6.4.0 (#115)

## What

Reverted the auto-labeler workflow from the release-drafter v7 autolabeler
sub-action back to the v6.4.0 main action with disable-releaser: true.

## Why

The v7 autolabeler sub-action hardcodes a strict check for the pull_request
event and rejects pull_request_target. Since this reusable workflow is called
from a pull_request_target trigger (required for labeling fork PRs with write
permissions), the autolabeler fails on every fork PR.

## Notes

- The v6 action uses GITHUB_TOKEN via env rather than the token input
- release-drafter v7 is still used in the release workflow; only the autolabeler is pinned to v6
- Consider opening an upstream issue on release-drafter to support pull_request_target in the v7 autolabeler

Signed-off-by: jmeridth <jmeridth@gmail.com>

Author: jmeridth

.github/workflows/auto-labeler.yaml

@@ -22,7 +22,9 @@ jobs:
         uses: step-security/harden-runner@58077d3c7e43986b6b15fba718e8ea69e387dfcc # v2.15.1
         with:
           egress-policy: audit
-      - uses: release-drafter/release-drafter/autolabeler@3a7fb5c85b80b1dda66e1ccb94009adbbd32fce3 # v7.0.0
+      - uses: release-drafter/release-drafter@6a93d829887aa2e0748befe2e808c66c0ec6e4c7 # v6.4.0
+        env:
+          GITHUB_TOKEN: ${{ secrets.github-token }}
         with:
           config-name: ${{ inputs.config-name }}
-          token: ${{ secrets.github-token }}
+          disable-releaser: true