do not panic on rebalance during server restarts (#19563)

This PR makes a feasible approach to handle all the scenarios
that we must face to avoid returning "panic."

Instead, we must return "errServerNotInitialized" when a
bucketMetadataSys.Get() is called, allowing the caller to
retry their operation and wait.

Bonus fix the way data-usage-cache stores the object.
Instead of storing usage-cache.bin with the bucket as
`.minio.sys/buckets`, the `buckets` must be relative
to the bucket `.minio.sys` as part of the object name.

Otherwise, there is no way to decommission entries at
`.minio.sys/buckets` and their final erasure set positions.

A bucket must never have a `/` in it. Adds code to read()
from existing data-usage.bin upon upgrade.

cmd/os-reliable.go

@@ -108,6 +108,16 @@ func reliableMkdirAll(dirPath string, mode os.FileMode, baseDir string) (err err
 			// Retry only for the first retryable error.
 			if osIsNotExist(err) && i == 0 {
 				i++
+				// Determine if os.NotExist error is because of
+				// baseDir's parent being present, retry it once such
+				// that the MkdirAll is retried once for the parent
+				// of dirPath.
+				// Because it is worth a retry to skip a different
+				// baseDir which is slightly higher up the depth.
+				nbaseDir := path.Dir(baseDir)
+				if baseDir != "" && nbaseDir != "" && nbaseDir != SlashSeparator {
+					baseDir = nbaseDir
+				}
 				continue
 			}
 		}