Update documentation for new graphql-java-tools example

spring-example/README.md

@@ -1,11 +1,16 @@
+This Spring Boot application contains examples for both a microservice using standard `graphql-java` and a microservice using `graphql-java-tools`; their code is separated under different subpackages and Spring profiles.
 
-To run the example, cd to the root of this project, then...
+To run the standard `graphql-java` example, `cd` to the root of this project, then...
 ```
 ## compile and install project including example and dependencies
 mvn install -Dgpg.skip
 ## start local webserver on 9000
 mvn -pl spring-example spring-boot:run
 ```
+To run the `graphql-java-tools` example, for the last step instead run:
+```
+mvn -pl spring-example spring-boot:run -Dspring-boot.run.profiles=graphql-java-tools
+```
 Now you can query your local graph:
 ```
 ## e.g.

spring-example/src/main/java/com/apollographql/federation/springexample/graphqljavatools/GraphQLJavaToolsConfiguration.java

@@ -16,6 +16,25 @@
 import org.springframework.context.annotation.Bean;
 import org.springframework.context.annotation.Configuration;
 
+/**
+ * Some tips for running with graphql-java-tools:
+ * 1. You may find that graphql-java-tools removes types that are only used by the federation
+ *    _entities field. This is because graphql-java-tools performs these optimizations before
+ *    federation-graphql-java-support can alter the schema to include the _entities field. The
+ *    simplest workaround here is to both (1) add the type to the {@link SchemaParserDictionary}
+ *    bean and to (2) set the {@link SchemaParserOptions.Builder} bean to have includeUnusedTypes
+ *    as true. However, the latter can't be done via properties.xml/yml since the getters/setters
+ *    aren't named according to the JavaBeans API specification (i.e. getFoo()/setFoo()). You can
+ *    work around this by providing the whole bean via code, or using a {@link BeanPostProcessor}
+ *    as shown below to customize the auto-configuration bean.
+ * 2. If you have an empty query type, you don't need to add a dummy field to your
+ *    {@link graphql.kickstart.tools.GraphQLQueryResolver} or to your schema string. However, you
+ *    do need to (1) declare an empty {@link graphql.kickstart.tools.GraphQLQueryResolver}, (2)
+ *    declare an empty "type Query" in your schema string, and (3) setup {@link SchemaTransformer}
+ *    as shown below. We could potentially modify the federation-graphql-java-support API to accept
+ *    {@link GraphQLSchema.Builder}s, but unfortunately those builders have no public getters, so
+ *    we wouldn't be able to copy their data.
+ */
 @Configuration
 public class GraphQLJavaToolsConfiguration {
     @Bean