[jvm-packages][xgboost4j-gpu] Support GPU dataframe and DeviceQuantileDMatrix (#7195)

Following classes are added to support dataframe in java binding:

- `Column` is an abstract type for a single column in tabular data.
- `ColumnBatch` is an abstract type for dataframe.

- `CuDFColumn` is an implementaiton of `Column` that consume cuDF column
- `CudfColumnBatch` is an implementation of `ColumnBatch` that consumes cuDF dataframe.

- `DeviceQuantileDMatrix` is the interface for quantized data.

The Java implementation mimics the Python interface and uses `__cuda_array_interface__` protocol for memory indexing.  One difference is on JVM package, the data batch is staged on the host as java iterators cannot be reset.

Co-authored-by: jiamingy <jm.yuan@outlook.com>

jvm-packages/xgboost4j-gpu/pom.xml

@@ -12,7 +12,24 @@
     <version>1.5.0-SNAPSHOT</version>
     <packaging>jar</packaging>
 
+    <properties>
+      <cudf.version>21.08.2</cudf.version>
+      <cudf.classifier>cuda11</cudf.classifier>
+    </properties>
+
     <dependencies>
+        <dependency>
+          <groupId>ai.rapids</groupId>
+          <artifactId>cudf</artifactId>
+          <version>${cudf.version}</version>
+          <classifier>${cudf.classifier}</classifier>
+          <scope>provided</scope>
+        </dependency>
+        <dependency>
+          <groupId>com.fasterxml.jackson.core</groupId>
+          <artifactId>jackson-databind</artifactId>
+          <version>2.10.5.1</version>
+        </dependency>
         <dependency>
             <groupId>org.apache.hadoop</groupId>
             <artifactId>hadoop-hdfs</artifactId>