Fix mojibake in signature: detect real encoding instead of assuming UTF-8

Outlook writes signature .htm files in the system ANSI code page
(windows-1251 on Russian Windows) without a BOM, but File.ReadAllText
defaults to UTF-8 for BOM-less files - mangling Cyrillic into
replacement characters. ReadSignatureHtml now checks BOM, then a
<meta charset> tag, then falls back to Encoding.Default (the actual
system code page) before ever assuming UTF-8.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

outlook-bulk-sender/BulkArchiveMailer.AddIn/UI/PreviewForm.cs

@@ -58,7 +58,7 @@ namespace BulkArchiveMailer.AddIn.UI
             {
                 if (!string.IsNullOrWhiteSpace(settings.SignatureHtmlPath))
                 {
-                    return File.Exists(settings.SignatureHtmlPath) ? File.ReadAllText(settings.SignatureHtmlPath) : null;
+                    return File.Exists(settings.SignatureHtmlPath) ? SignatureReader.ReadSignatureHtml(settings.SignatureHtmlPath) : null;
                 }
 
                 return SignatureReader.TryAutoDetectSignatureHtml(outlookApplication);