Wrong parsing of text/html with boundary

[kadukf-outreach]

What I did:
Parsing a message which has text/html part but it contains also "boundary" marker. I know, this is an strange case but we have the messages with it.

IMO,

enmime/part.go

Line 404 in 1f2f4b0

if p.Boundary == "" {

we could check if the content type with "detectTextHeader", wdyt?

Or reporting it with error and providing content to the part. IMO the part is wrongly classified.

What I expected:
To have HTML text loaded from that part.

What I got:
Empty string + I cant access the body of that part :(

Release or branch I am using:
master
(Please attach a sample message if you feel it will help reproduce the issue)

------=_NextPart_001_0564_01D7A4BE.5DCBAE50

Content-Type: text/html;

boundary="_005_CO1PR15MB50972F18B0F86496F538DA40A1D49CO1PR15MB5097namp_";

charset="us-ascii";

type="multipart/alternative"

Content-Transfer-Encoding: quoted-printable

[jhillyerd]

Apologies for slow response. I agree, using detectTextHeader to not treat it at multipart, and adding a new error to error.go would be a good approach.