docker: use correct effective cpuset filename on legacy cgroups v1 systems

[shoenig]

Fixes #20293

[shoenig]

Spot Check)

➜ make dev && sudo nomad agent -dev

job 1 (using resources.cpu)

job "job1" {

  group "cache" {
    network {
      port "db" {
        to = 6379
      }
    }

    task "redis" {
      driver = "docker"

      config {
        image          = "redis:7"
        ports          = ["db"]
        auth_soft_fail = true
      }

      resources {
        cpu    = 500
        memory = 256
      }
    }
  }
}

job 2 (using resources.cores)

job "job2" {

  group "cache" {
    network {
      port "db" {
        to = 6379
      }
    }

    task "redis" {
      driver = "docker"

      config {
        image          = "redis:7"
        ports          = ["db"]
        auth_soft_fail = true
      }

      resources {
        cores = 3
        memory = 256
      }
    }
  }
}

➜ nomad job run job1.hcl

initial value is all cpus as expected

➜ cat /sys/fs/cgroup/cpuset/docker/d047294cd29a0c0abd0d76509b837acedf0aff7391acc4d5c8232ffb2dfa68d5/cpuset.cpus
0-11

➜ nomad job run job2.hcl

cpuset has been updated to exclude cores reserved for job2

➜ cat /sys/fs/cgroup/cpuset/docker/d047294cd29a0c0abd0d76509b837acedf0aff7391acc4d5c8232ffb2dfa68d5/cpuset.cpus
3-11

[github-actions]

I'm going to lock this pull request because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.