(export) Filter non-HTML documents in exporters

Add a check to ensure only documents with "text/html" content type are processed in FeedExporter, AtagExporter, and TermFrequencyExporter. This prevents non-HTML documents from being parsed and helps maintain data consistency and keep the memory usage down.

code/processes/export-task-process/java/nu/marginalia/extractor/AtagExporter.java

@@ -89,6 +89,8 @@ public class AtagExporter implements ExporterIf {
                 continue;
             if (null == doc.documentBody)
                 continue;
+            if (!doc.contentType.toLowerCase().startsWith("text/html"))
+                continue;
 
             var baseUrl = new EdgeUrl(doc.url);
             var parsed = Jsoup.parse(doc.documentBody);

code/processes/export-task-process/java/nu/marginalia/extractor/FeedExporter.java

@@ -83,6 +83,8 @@ public class FeedExporter implements ExporterIf {
                 continue;
             if (null == doc.documentBody)
                 continue;
+            if (!doc.contentType.toLowerCase().startsWith("text/html"))
+                continue;
 
             var baseUrl = new EdgeUrl(doc.url);
             var parsed = Jsoup.parse(doc.documentBody);

code/processes/export-task-process/java/nu/marginalia/extractor/TermFrequencyExporter.java

@@ -111,7 +111,7 @@ public class TermFrequencyExporter implements ExporterIf {
 
                 if (!(stream.next() instanceof CrawledDocument doc)) continue;
                 if (doc.documentBody == null) continue;
-                if (!doc.contentType.startsWith("text/html"))
+                if (!doc.contentType.toLowerCase().startsWith("text/html"))
                     continue;
 
                 docCount.incrementAndGet();