fix(appsec): use machine default charset as fallback in MultipartContentDecoder

Replaces hardcoded UTF-8 (no-charset default) and ISO-8859-1 (fallback)
with Charset.defaultCharset() in both cases, per reviewer feedback.

Author: jandro996

internal-api/src/main/java/datadog/trace/api/http/MultipartContentDecoder.java

@@ -4,15 +4,14 @@
 import java.nio.charset.CharacterCodingException;
 import java.nio.charset.Charset;
 import java.nio.charset.CodingErrorAction;
-import java.nio.charset.StandardCharsets;
 import java.util.Locale;
 
 /** Decodes multipart file content bytes to String using the per-part Content-Type charset. */
 public final class MultipartContentDecoder {
 
   public static String decodeBytes(byte[] buf, int length, String contentType) {
     Charset charset = extractCharset(contentType);
-    if (charset == null) charset = StandardCharsets.UTF_8;
+    if (charset == null) charset = Charset.defaultCharset();
     try {
       return charset
           .newDecoder()
@@ -21,7 +20,7 @@ public static String decodeBytes(byte[] buf, int length, String contentType) {
           .decode(ByteBuffer.wrap(buf, 0, length))
           .toString();
     } catch (CharacterCodingException e) {
-      return new String(buf, 0, length, StandardCharsets.ISO_8859_1);
+      return new String(buf, 0, length, Charset.defaultCharset());
     }
   }
 

internal-api/src/test/groovy/datadog/trace/api/http/MultipartContentDecoderTest.groovy

@@ -1,5 +1,6 @@
 package datadog.trace.api.http
 
+import java.nio.charset.Charset
 import spock.lang.Specification
 
 class MultipartContentDecoderTest extends Specification {
@@ -13,7 +14,7 @@ class MultipartContentDecoderTest extends Specification {
     MultipartContentDecoder.decodeBytes(bytes, bytes.length, 'text/plain; charset=UTF-8') == text
   }
 
-  void 'decodeBytes falls back to UTF-8 when Content-Type has no charset'() {
+  void 'decodeBytes falls back to machine default when Content-Type has no charset'() {
     given:
     def text = 'hello world'
     byte[] bytes = text.getBytes('UTF-8')
@@ -22,7 +23,7 @@ class MultipartContentDecoderTest extends Specification {
     MultipartContentDecoder.decodeBytes(bytes, bytes.length, 'text/plain') == text
   }
 
-  void 'decodeBytes falls back to UTF-8 when Content-Type is null'() {
+  void 'decodeBytes falls back to machine default when Content-Type is null'() {
     given:
     def text = 'hello world'
     byte[] bytes = text.getBytes('UTF-8')
@@ -31,13 +32,13 @@ class MultipartContentDecoderTest extends Specification {
     MultipartContentDecoder.decodeBytes(bytes, bytes.length, null) == text
   }
 
-  void 'decodeBytes falls back to ISO-8859-1 when bytes are invalid for declared charset'() {
+  void 'decodeBytes falls back to machine default when bytes cannot be decoded with default charset'() {
     given:
-    // 0xE9 is 'é' in ISO-8859-1 but an invalid lone UTF-8 byte
     byte[] bytes = 'café'.getBytes('ISO-8859-1')
+    String expected = new String(bytes, Charset.defaultCharset())
 
     expect:
-    MultipartContentDecoder.decodeBytes(bytes, bytes.length, null) == 'café'
+    MultipartContentDecoder.decodeBytes(bytes, bytes.length, null) == expected
   }
 
   void 'decodeBytes uses declared ISO-8859-1 charset'() {
@@ -62,13 +63,14 @@ class MultipartContentDecoderTest extends Specification {
     MultipartContentDecoder.decodeBytes(new byte[16], 0, null) == ''
   }
 
-  void 'decodeBytes falls back to ISO-8859-1 when declared charset cannot decode the bytes'() {
+  void 'decodeBytes falls back to machine default when declared charset cannot decode the bytes'() {
     given:
     // bytes are ISO-8859-1 encoded but Content-Type explicitly declares UTF-8
     byte[] bytes = 'café'.getBytes('ISO-8859-1')
+    String expected = new String(bytes, Charset.defaultCharset())
 
     expect:
-    MultipartContentDecoder.decodeBytes(bytes, bytes.length, 'text/plain; charset=UTF-8') == 'café'
+    MultipartContentDecoder.decodeBytes(bytes, bytes.length, 'text/plain; charset=UTF-8') == expected
   }
 
   void 'extractCharset returns null for null contentType'() {