Optimize XContent Object Parsers (#78813)

Optimize object parsers a little by extracting cold paths, removing some unnecessary
lambda wrapping and some other small things.
Also, fixed a very expensive use of these APIs in Phase moving from a very hot stream
instantiation to a standard loop.

Author: original-brownbear

libs/x-content/src/main/java/org/elasticsearch/xcontent/AbstractObjectParser.java

@@ -156,7 +156,7 @@ public <T> void declareField(BiConsumer<Value, T> consumer, CheckedFunction<XCon
     }
 
     public <T> void declareObject(BiConsumer<Value, T> consumer, ContextParser<Context, T> objectParser, ParseField field) {
-        declareField(consumer, (p, c) -> objectParser.parse(p, c), field, ValueType.OBJECT);
+        declareField(consumer, objectParser, field, ValueType.OBJECT);
     }
 
     /**
@@ -240,7 +240,7 @@ public void declareBoolean(BiConsumer<Value, Boolean> consumer, ParseField field
     }
 
     public <T> void declareObjectArray(BiConsumer<Value, List<T>> consumer, ContextParser<Context, T> objectParser, ParseField field) {
-        declareFieldArray(consumer, (p, c) -> objectParser.parse(p, c), field, ValueType.OBJECT_ARRAY);
+        declareFieldArray(consumer, objectParser, field, ValueType.OBJECT_ARRAY);
     }
 
     /**

libs/x-content/src/main/java/org/elasticsearch/xcontent/ConstructingObjectParser.java

@@ -426,7 +426,7 @@ private class Target {
         /**
          * The parse context that is used for this invocation. Stored here so that it can be passed to the {@link #builder}.
          */
-        private Context context;
+        private final Context context;
 
         /**
          * How many of the constructor parameters have we collected? We keep track of this so we don't have to count the