internal-api: accept tab as boundary before charset= in MultipartContentDecoder

RFC 7230 optional whitespace (OWS) allows both space and horizontal tab.
extractCharset() only accepted space, so "type;\tcharset=X" was treated as
no-charset-declared, causing MultipartHelper to silently override the
client's declared charset with UTF-8.

Author: jandro996

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

@@ -43,8 +43,12 @@ public static Charset extractCharset(String contentType) {
     while (true) {
       int idx = indexOfIgnoreAsciiCase(contentType, "charset=", searchFrom);
       if (idx < 0) return null;
-      // Require a parameter boundary before "charset=" so "xcharset=..." is not matched
-      if (idx == 0 || contentType.charAt(idx - 1) == ';' || contentType.charAt(idx - 1) == ' ') {
+      // Require a parameter boundary before "charset=" so "xcharset=..." is not matched.
+      // RFC 7230 optional whitespace (OWS) allows both space and horizontal tab.
+      if (idx == 0
+          || contentType.charAt(idx - 1) == ';'
+          || contentType.charAt(idx - 1) == ' '
+          || contentType.charAt(idx - 1) == '\t') {
         int nameStart = idx + 8;
         int end = contentType.length();
         for (int i = nameStart; i < end; i++) {

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

@@ -111,6 +111,18 @@ void extractCharsetFromStandardContentType(String contentType, String expectedCh
     assertEquals(expectedCharset, MultipartContentDecoder.extractCharset(contentType).name());
   }
 
+  @ParameterizedTest
+  @ValueSource(
+      strings = {
+        "text/plain;\tcharset=ISO-8859-1",
+        "text/plain; \tcharset=ISO-8859-1",
+        "text/plain;\t charset=ISO-8859-1"
+      })
+  void extractCharsetAcceptsTabAsParameterBoundary(String contentType) {
+    // RFC 7230 optional whitespace (OWS) allows both space and horizontal tab before a parameter
+    assertEquals("ISO-8859-1", MultipartContentDecoder.extractCharset(contentType).name());
+  }
+
   @Test
   void extractCharsetIgnoresSubstringMatchInParameterName() {
     // "xcharset=UTF-16" must not match; the real "charset=UTF-8" that follows must be used