avoid disk monitoring leaks under various conditions (#18777)

- HealFormat() was leaking healthcheck goroutines for
  disks, we are only interested in enabling healthcheck
  for the newly formatted disk, not for existing disks.

- When disk is a root-disk a random disk monitor was
  leaking while we ignored the drive.

- When loading the disk for each erasure set, we were
  leaking goroutines for the prepare-storage.go disks
  which were replaced via the globalLocalDrives slice

- avoid disk monitoring utilizing health tokens that
  would cause exhaustion in the tokens, prematurely
  which were meant for incoming I/O. This is ensured
  by avoiding writing O_DIRECT aligned buffer instead
  write 2048 worth of content only as O_DSYNC, which is
  sufficient.

cmd/xl-storage.go

@@ -715,11 +715,7 @@ func (s *xlStorage) DiskInfo(_ context.Context, _ bool) (info DiskInfo, err erro
 	s.diskInfoCache.Once.Do(func() {
 		s.diskInfoCache.TTL = time.Second
 		s.diskInfoCache.Update = func() (interface{}, error) {
-			dcinfo := DiskInfo{
-				RootDisk:  s.rootDisk,
-				MountPath: s.drivePath,
-				Endpoint:  s.endpoint.String(),
-			}
+			dcinfo := DiskInfo{}
 			di, err := getDiskInfo(s.drivePath)
 			if err != nil {
 				return dcinfo, err
@@ -748,6 +744,10 @@ func (s *xlStorage) DiskInfo(_ context.Context, _ bool) (info DiskInfo, err erro
 	if v != nil {
 		info = v.(DiskInfo)
 	}
+
+	info.RootDisk = s.rootDisk
+	info.MountPath = s.drivePath
+	info.Endpoint = s.endpoint.String()
 	info.Scanning = atomic.LoadInt32(&s.scanning) == 1
 	return info, err
 }