[CI] Add dispatch job to leverage dynamic devices

[Yikun]

What this PR does / why we need it?

Add dispatch job to leverage jobs to dynamic devices include 2 stage as below:

The dispatch job will spend extra about 10s * parallel number + 30s time to wait other job launch container and release lock.

• Stage 1: Acquire lock
  add a dispatch job, this job use lockfile to acquire locks and then get device number dynamically
• Stage 2.1: Launch container with dynamic device
  pass the device number via output and start the container job with dynamic device
• Stage 2.2: Release lock
  once the job started, release the lock.

In the backend, we use multiple path to setup multiple self host runners as load balancer:

$ pwd
/home/action
$ ll | grep actions
drwx------   6 action action 4096 Mar  7 08:55 actions-runner-01
drwx------   6 action action 4096 Mar  7 08:55 actions-runner-02
drwx------   6 action action 4096 Mar  7 08:55 actions-runner-03
drwx------   6 action action 4096 Mar  7 08:56 actions-runner-04
drwx------   4 action action 4096 Jan 24 22:08 actions-runner-05
drwx------   4 action action 4096 Jan 24 22:08 actions-runner-06

adduser -G docker action
su action
pip3 install docker prettytable
sudo yum install procmail

Does this PR introduce any user-facing change?

NO

How was this patch tested?

• CI passed
• E2E test manully, triggered 3 jobs in parallel:
  • 1st job dispatch to /dev/davinci2.
  • 2nd job dispatch to /dev/davinci3
  • 3rd job dispatch to /dev/davinci4
    

[Yikun]

See also bug fix: 663dca7 to remove lock @pkking

  post_cleanup:
    name: vLLM Ascend test (post-cleanup)
    needs: [test]
    runs-on: ascend-ci-arm64  # actionlint-ignore: runner-label
    if: always()
    steps:
      - name: Remove dispatch lock if exists
        run: |
          if [ -f "/tmp/dispatch.lock" ]; then
            rm -f "/tmp/dispatch.lock"
          fi