[java] Fix the logic of generating TaskID (#2747)

## What do these changes do?
Because the logic of generating `TaskID` in java is different from python's, there are many tests fail when we change the `Ray Core` code.
In this change,  I rewrote the logic of generating `TaskID` in java which is the same as the python's.

In java, we call the native method `_generateTaskId()` to generate a `TaskID` which is also used in python. We change `computePutId()`'s logic too.

## Related issue number
[#2608](https://github.com/ray-project/ray/issues/2608)

java/api/src/main/java/org/ray/api/UniqueID.java

@@ -7,7 +7,7 @@ import java.util.Random;
 import javax.xml.bind.DatatypeConverter;
 
 /**
- * Unique ID for task, worker, function...
+ * Unique ID for task, worker, function.
  */
 public class UniqueID implements Serializable {
 
@@ -42,7 +42,8 @@ public class UniqueID implements Serializable {
 
   public UniqueID(byte[] id) {
     if (id.length != LENGTH) {
-      throw new IllegalArgumentException("Illegal argument: " + id.toString());
+      throw new IllegalArgumentException("Illegal argument for UniqueID, expect " + LENGTH
+          + " bytes, but got " + id.length + " bytes.");
     }
 
     this.id = id;
@@ -86,11 +87,6 @@ public class UniqueID implements Serializable {
   }
 
   public boolean isNil() {
-    for (byte b : id) {
-      if (b != (byte) 0xFF) {
-        return false;
-      }
-    }
-    return true;
+    return this.equals(NIL);
   }
 }