The big ClassLoader change

This changes the way Quarkus ClassLoading works,
to allow for isolated class loaders.

It also unifies how Quarkus is launched, so every
different mode we support uses the same mechanism
for both curation and launch.

Tests are now run in an isolated ClassLoader, which
means that a proxy is created that runs the tests
from within the isolated ClassLoader. This currently
has a quirk where @BeforeAll methods are run twice,
which will be fixed in the next JUnit release. This
can be worked around by using @QuarkusBeforeAll.

extensions/tika/deployment/src/main/java/io/quarkus/tika/deployment/TikaProcessor.java

@@ -16,6 +16,7 @@
 import org.apache.tika.detect.EncodingDetector;
 import org.apache.tika.parser.Parser;
 
+import io.quarkus.arc.deployment.AdditionalBeanBuildItem;
 import io.quarkus.arc.deployment.BeanContainerBuildItem;
 import io.quarkus.deployment.Capabilities;
 import io.quarkus.deployment.annotations.BuildProducer;
@@ -32,6 +33,7 @@
 import io.quarkus.tika.TikaParseException;
 import io.quarkus.tika.runtime.TikaConfiguration;
 import io.quarkus.tika.runtime.TikaParserParameter;
+import io.quarkus.tika.runtime.TikaParserProducer;
 import io.quarkus.tika.runtime.TikaRecorder;
 
 public class TikaProcessor {
@@ -52,6 +54,11 @@ public class TikaProcessor {
 
     private TikaConfiguration config;
 
+    @BuildStep
+    AdditionalBeanBuildItem beans() {
+        return AdditionalBeanBuildItem.builder().addBeanClasses(TikaParserProducer.class).build();
+    }
+
     @BuildStep
     @Record(ExecutionTime.STATIC_INIT)
     TikaParsersConfigBuildItem initializeTikaParser(BeanContainerBuildItem beanContainer, TikaRecorder recorder)