fix: MIME Decoding corrupts non-ASCII characters in Base64-encoded words

fromBase64() defaults to returning a UTF-8 decoded string, which is then
passed to codepage.utils.decode() that treats each char code as a raw
byte.
For multi-byte UTF-8 characters, this double-decoding produces garbage
(e.g. "café" becomes "caf退").

Pass returnType="byteArray" so codepage receives raw bytes and performs
the single correct UTF-8 decode.

Closes #2280

src/core/operations/MIMEDecoding.mjs

@@ -87,7 +87,7 @@ class MIMEDecoding extends Operation {
             end = cur + j + "?=".length;
 
             if (encoding.toLowerCase() === "b") {
-                text = fromBase64(text);
+                text = fromBase64(text, undefined, "byteArray");
             } else if (encoding.toLowerCase() === "q") {
                 text = this.parseQEncodedWord(text);
             } else {

tests/operations/tests/MIMEDecoding.mjs

@@ -75,6 +75,39 @@ TestRegister.addTests([
             }
         ]
     },
+    {
+        name: "UTF-8 Base64 non-ASCII",
+        input: "Subject: =?UTF-8?B?Y2Fmw6k=?=",
+        expectedOutput: "Subject: café",
+        recipeConfig: [
+            {
+                "op": "MIME Decoding",
+                "args": []
+            }
+        ]
+    },
+    {
+        name: "UTF-8 Base64 multibyte CJK",
+        input: "Subject: =?UTF-8?B?5pel5pys6Kqe?=",
+        expectedOutput: "Subject: 日本語",
+        recipeConfig: [
+            {
+                "op": "MIME Decoding",
+                "args": []
+            }
+        ]
+    },
+    {
+        name: "UTF-8 Base64 ASCII-only",
+        input: "Subject: =?UTF-8?B?aGVsbG8=?=",
+        expectedOutput: "Subject: hello",
+        recipeConfig: [
+            {
+                "op": "MIME Decoding",
+                "args": []
+            }
+        ]
+    },
     {
         name: "ISO Decoding",
         input: "From: =?US-ASCII?Q?Keith_Moore?= <moore@cs.utk.edu>\nTo: =?ISO-8859-1?Q?Keld_J=F8rn_Simonsen?= <keld@dkuug.dk>\nCC: =?ISO-8859-1?Q?Andr=E9?= Pirard <PIRARD@vm1.ulg.ac.be>\nSubject: =?ISO-8859-1?B?SWYgeW91IGNhbiByZWFkIHRoaXMgeW8=?=\n=?ISO-8859-2?B?dSB1bmRlcnN0YW5kIHRoZSBleGFtcGxlLg==?=",