Fix several bugs with handling e-mail files

- When converting from html->pdf, the wkhtmltopdf program exits with
  errors if the document contains invalid links. The content is now
  cleaned before handed to wkhtmltopdf.
- Update emil library which fixes a bug when reading mails without
  explicit transfer encoding (8bit)
- Add a info header to converted mails

modules/common/src/main/scala/docspell/common/Binary.scala

@@ -1,5 +1,6 @@
 package docspell.common
 
+import cats.effect._
 import fs2.{Chunk, Pipe, Stream}
 import java.nio.charset.Charset
 import java.nio.charset.StandardCharsets
@@ -42,6 +43,9 @@ object Binary {
       util.decode[F](cs)
     }
 
+  def loadAllBytes[F[_]: Sync](data: Stream[F, Byte]): F[ByteVector] =
+    data.chunks.map(_.toByteVector).compile.fold(ByteVector.empty)((r, e) => r ++ e)
+
   // This is a copy from org.http4s.util
   // Http4s is licensed under the Apache License 2.0
   private object util {
@@ -85,5 +89,6 @@ object Binary {
       if (chunk.size >= 3 && chunk.take(3) == utf8Bom) {
         chunk.drop(3)
       } else chunk
+
   }
 }