[FLINK-19981][core][table] Add name-based field mode for Row

This adds a name-based field mode to the Row class. A row can
 operate in 3 different modes: name-based, position-based, or
a hybrid of both when leaving the Flink runtime. It simplifies
the handling of large rows (possibly with hundreds of fields)
and will make it easier to switch between DataStream API and
Table API.

See the documentation of the Row class for more information.

This closes apache#14420.

flink-core/src/main/java/org/apache/flink/api/java/typeutils/RowTypeInfo.java

@@ -32,6 +32,7 @@
 import java.util.Arrays;
 import java.util.Collections;
 import java.util.HashSet;
+import java.util.LinkedHashMap;
 import java.util.List;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
@@ -253,7 +254,11 @@ public TypeSerializer<Row> createSerializer(ExecutionConfig config) {
         for (int i = 0; i < len; i++) {
             fieldSerializers[i] = types[i].createSerializer(config);
         }
-        return new RowSerializer(fieldSerializers);
+        final LinkedHashMap<String, Integer> positionByName = new LinkedHashMap<>();
+        for (int i = 0; i < fieldNames.length; i++) {
+            positionByName.put(fieldNames[i], i);
+        }
+        return new RowSerializer(fieldSerializers, positionByName);
     }
 
     @Override
@@ -308,7 +313,7 @@ public TypeSerializer<Row> createLegacySerializer(ExecutionConfig config) {
         for (int i = 0; i < len; i++) {
             fieldSerializers[i] = types[i].createSerializer(config);
         }
-        return new RowSerializer(fieldSerializers, true);
+        return new RowSerializer(fieldSerializers, null, true);
     }
 
     /** Tests whether an other object describes the same, schema-equivalent row information. */