fix: fix for quote-printed utf-8 header with quotes

[dmytrokasianenko-outreach]

This PR fixes decoding addresses with quotes.
For example, "=?UTF-8?Q?Miros=C5=82aw_Marczak?=" should be decoded as Mirosław Marczak.

[jhillyerd]

I think this may revert the fix for #218 by decoding the 2047 words too early.

[jhillyerd]

Thinking about how to get better test coverage for this, without introducing a bunch more file based envelope tests:

We could copy https://github.com/jhillyerd/enmime/blob/master/header_test.go#L60 (TestParseAddressListResult) to envelope_test.go, then modify it to create a minimal message, putting the input values on the To: line, parsing into an envelope and then checking what the To field contains.

[dmytrokasianenko-outreach]

You are right, it was breaking the the fix for #218.

I've changed solution by trying to additionally decode address.Name with mime.WordDecoder. It works fine for the test cases, however I'm not sure it is a valid one. What do you think?

[dmytrokasianenko-outreach]

I've added a fix for ParseAddressList to be able to decode different charsets

[jhillyerd]

I'll try and take a deeper look at the address.Name decoding in the next few days, thanks

[jhillyerd]

I think this is a good, thanks for your work on this!

[dmytrokasianenko-outreach]

thank you for all your work!