2. Add function `GetMultiTenantServicePrincipalToken` to fetch multi-tenant service principal token, which will be used by Azure VM/VMSS Clients in this feature.

			return psi, errSRPeerResp(fmt.Errorf("unable to create admin client for %s: %w", v.Name, err))
		}

		info, err := admClient.ServerInfo(ctx)
		if err != nil {
			return psi, errSRPeerResp(fmt.Errorf("unable to fetch server info for %s: %w", v.Name, err))
		}

		s3Client, err := getS3Client(v)
		if err != nil {
			return psi, errSRPeerResp(fmt.Errorf("unable to create s3 client for %s: %w", v.Name, err))
		}

- Fixed a regression in default 1.27 configurations in kube-apiserver: fixed the AggregatedDiscoveryEndpoint feature (beta in 1.27+) to successfully fetch discovery information from aggregated API servers that do not check `Accept` headers when serving the `/apis` endpoint ([#120359](https://github.com/kubernetes/kubernetes/pull/120359), [@Jefftree](https://github.com/Jefftree)) [SIG API Machinery]

* Fix cluster autoscaler addon permissions so it can access batch/job. ([#69858](https://github.com/kubernetes/kubernetes/pull/69858), [@losipiuk](https://github.com/losipiuk))

        if (!closed) {
          put(closeable, executor);
          return;
        }
      }
      closeQuietly(closeable, executor);
    }

    /**
     * Returns a latch that reaches zero when this objects' deferred closeables have been closed.
     */
    CountDownLatch whenClosedCountDown() {
      if (closed) {
        return new CountDownLatch(0);
      }

- Updated the `v1` credential provider configuration to include the `tokenAttributes.cacheType` field. This field is required and must be set to either `ServiceAccount` or `Token` when configuring a provider that uses a service account to fetch registry credentials. ([#132617](https://github.com/kubernetes/kubernetes/pull/132617), [@aramase](https://github.com/aramase)) [SIG Auth, Node and Testing]

		Users:    make(map[string]set.StringSet),
		Groups:   set.CreateStringSet(q.Groups...),
		Policies: set.CreateStringSet(q.Policy...),
	}

	if ldap {
		// Validate and normalize users, then fetch and normalize their groups
		// Also include unvalidated users for backward compatibility.
		for _, user := range q.Users {
			lookupRes, actualGroups, _ := sys.LDAPConfig.GetValidatedDNWithGroups(user)

* Fix cluster autoscaler addon permissions so it can access batch/job. ([#69858](https://github.com/kubernetes/kubernetes/pull/69858), [@losipiuk](https://github.com/losipiuk))

- Fixed a regression in the default `v1.27` configurations in `kube-apiserver`: fixed the `AggregatedDiscoveryEndpoint` feature (`beta` in `v1.27+`) to successfully fetch discovery information from aggregated API servers that do not check `Accept` headers when serving the `/apis` endpoint. ([#119870](https://github.com/kubernetes/kubernetes/pull/119870), [@Jefftree](https://github.com/Jefftree))

		// Optimization for Spark/Hadoop workload where spark sends a garbage
		// request of this kind
		//
		// GET /testbucket/?list-type=2&delimiter=%2F&max-keys=2&prefix=parquet%2F_SUCCESS%2F&fetch-owner=false
		//
		// Here spark is expecting that the List() return empty instead, so from MinIO's point
		// of view if we simply do a GetObjectInfo() on this prefix by treating it as an object