MAPREDUCE-6820. Fix dead links in Job relevant classes. Contributed by Yiqun Lin.

(cherry picked from commit 92a8917ca5a25bf6bee75acbb554aae87766a7e7)
(cherry picked from commit 1e1ab9bc1bde48593eceafb7380d7501d4bbc200)

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/JobConf.java

@@ -1319,7 +1319,7 @@ public class JobConf extends Configuration {
    * bytes, of input files. However, the {@link FileSystem} blocksize of the 
    * input files is treated as an upper bound for input splits. A lower bound 
    * on the split size can be set via 
-   * <a href="{@docRoot}/../mapred-default.html#mapreduce.input.fileinputformat.split.minsize">
+   * <a href="{@docRoot}/../hadoop-mapreduce-client/hadoop-mapreduce-client-core/mapred-default.xml#mapreduce.input.fileinputformat.split.minsize">
    * mapreduce.input.fileinputformat.split.minsize</a>.</p>
    *  
    * <p>Thus, if you expect 10TB of input data and have a blocksize of 128MB, 

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/Mapper.java

@@ -147,7 +147,7 @@ public interface Mapper<K1, V1, K2, V2> extends JobConfigurable, Closeable {
    * takes significant amount of time to process individual key/value
    * pairs, this is crucial since the framework might assume that the task has 
    * timed-out and kill that task. The other way of avoiding this is to set 
-   * <a href="{@docRoot}/../mapred-default.html#mapreduce.task.timeout">
+   * <a href="{@docRoot}/../hadoop-mapreduce-client/hadoop-mapreduce-client-core/mapred-default.xml#mapreduce.task.timeout">
    * mapreduce.task.timeout</a> to a high-enough value (or even zero for no 
    * time-outs).</p>
    * 

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/Reducer.java

@@ -187,7 +187,7 @@ public interface Reducer<K2, V2, K3, V3> extends JobConfigurable, Closeable {
    * takes a significant amount of time to process individual key/value 
    * pairs, this is crucial since the framework might assume that the task has 
    * timed-out and kill that task. The other way of avoiding this is to set 
-   * <a href="{@docRoot}/../mapred-default.html#mapreduce.task.timeout">
+   * <a href="{@docRoot}/../hadoop-mapreduce-client/hadoop-mapreduce-client-core/mapred-default.xml#mapreduce.task.timeout">
    * mapreduce.task.timeout</a> to a high-enough value (or even zero for no 
    * time-outs).</p>
    *