[MINOR] docs: update document and build script for Hadoop-3.2 (#912)

### What changes were proposed in this pull request?

- Update document, including the supported mapReduce version and build command.
- Update build_distribution.sh. We can't recognize the hadoop version from final package name for now. We should update the final package name.

README.md

@@ -70,7 +70,7 @@ Currently supports Spark 2.3.x, Spark 2.4.x, Spark 3.0.x, Spark 3.1.x, Spark 3.2
 Note: To support dynamic allocation, the patch(which is included in patch/spark folder) should be applied to Spark
 
 ## Supported MapReduce Version
-Currently supports the MapReduce framework of Hadoop 2.8.5
+Currently supports the MapReduce framework of Hadoop 2.8.5, Hadoop 3.2.1
 
 ## Building Uniffle
 > note: currently Uniffle requires JDK 1.8 to build, adding later JDK support is on our roadmap.
@@ -96,6 +96,14 @@ Build against Spark 3.2.0
 
     mvn -DskipTests clean package -Pspark3.2.0
 
+Build against Hadoop 2.8.5
+
+    mvn -DskipTests clean package -Pmr,hadoop2.8
+
+Build against Hadoop 3.2.1
+
+    mvn -DskipTests clean package -Pmr,hadoop3.2
+
 To package the Uniffle, run:
 
     ./build_distribution.sh
@@ -108,6 +116,10 @@ Package against Spark 3.2.0, run:
 
     ./build_distribution.sh --spark3-profile 'spark3.2.0'
 
+Package will build against Hadoop 2.8.5 in default. If you want to build package against Hadoop 3.2.1, run:
+
+    ./build_distribution.sh --hadoop-profile 'hadoop3.2'
+
 rss-xxx.tgz will be generated for deployment
 
 ## Deploy

build_distribution.sh

@@ -48,6 +48,7 @@ SPARK2_MVN_OPTS=""
 SPARK3_PROFILE_ID="spark3"
 SPARK3_MVN_OPTS=""
 HADOOP_PROFILE_ID="hadoop2.8"
+NAME=none
 while (( "$#" )); do
   case $1 in
     --spark2-profile)
@@ -70,6 +71,10 @@ while (( "$#" )); do
       HADOOP_PROFILE_ID=$2
       shift
       ;;
+    --name)
+      NAME="$2"
+      shift
+      ;;
     --help)
       exit_with_usage
       ;;
@@ -122,6 +127,10 @@ SPARK3_VERSION=$("$MVN" help:evaluate -Dexpression=spark.version -P$SPARK3_PROFI
     | grep -v "WARNING"\
     | tail -n 1)
 
+if [ "$NAME" == "none" ]; then
+  NAME=$HADOOP_PROFILE_ID
+fi
+
 echo "RSS version is $VERSION"
 
 export MAVEN_OPTS="${MAVEN_OPTS:--Xmx2g -XX:ReservedCodeCacheSize=1g}"
@@ -139,7 +148,7 @@ echo -e "\$ ${BUILD_COMMAND[@]}\n"
 
 
 # Make directories
-DISTDIR="rss-$VERSION"
+DISTDIR="rss-$VERSION-$NAME"
 rm -rf "$DISTDIR"
 mkdir -p "${DISTDIR}/jars"
 echo "RSS ${VERSION}${GITREVSTRING} built for Hadoop ${HADOOP_VERSION} Spark2 ${SPARK2_VERSION} Spark3 ${SPARK3_VERSION}" >"${DISTDIR}/RELEASE"
@@ -210,6 +219,6 @@ cp $MR_CLIENT_JAR $MR_CLIENT_JAR_DIR
 cp -r bin $DISTDIR
 cp -r conf $DISTDIR
 
-rm -rf "rss-$VERSION.tgz"
-tar czf "rss-$VERSION.tgz" $DISTDIR
+rm -rf "rss-$VERSION-$NAME.tgz"
+tar czf "rss-$VERSION-$NAME.tgz" $DISTDIR
 rm -rf $DISTDIR