List all co-authors

Using the latest biblatex 3.3 and biber 2.4 development versions from Sourceforge. New mapping functionality allows loops and creating new entries and so it is possible to split up name fields into dedicated entrytypes for correct printing:

\documentclass{article}
\usepackage{filecontents}
% Declare the coauthor name field
\begin{filecontents}{\jobname.dbx}
\DeclareDatamodelFields[type=list, datatype=name, label=true]{coauthor}
\DeclareDatamodelEntrytypes{coauthor}
\DeclareDatamodelEntryfields[coauthor]{coauthor}
\end{filecontents}
\begin{filecontents}{\jobname.bib}
@ARTICLE{test1,
 AUTHOR = {Arthur Smith and James Jones and Bill Smiley and Jake Wrath},
 JOURNALTITLE = {A Journal},
 YEAR   = {2010}
}

@ARTICLE{test2,
 AUTHOR = {Philippa Smyth and Arthur Smith and Bill Likely},
 JOURNALTITLE = {Another Journal},
 YEAR   = {2011}
}

@ARTICLE{test3,
 AUTHOR = {Arthur Smith},
 JOURNALTITLE = {The Best Journal},
 YEAR   = {2013}
}

@ARTICLE{test4,
 AUTHOR = {Philippa Smyth and Arthur Smith and James Jones},
 JOURNALTITLE = {Again Another Journal},
 YEAR   = {2014}
}
\end{filecontents}
\usepackage[style=authoryear,datamodel=\jobname]{biblatex}
\addbibresource{\jobname.bib}

% 1. First, check if the author list contain me, here, "Arthur Smith"
% 2. Copy the author field to a temporary field "coauthortext". We don't need
%    datamodel declarations for this as this will be deleted after use and
%    will never get anywhere near code that queries the data model.
% 3. Remove me (Arthur Smith) from this temporary copy of the author list
% 4. Turn the resulting field into a comma-separated list
% 5. Loop over this comma-separated list, creating new entries of type
%    "coauthor" each with a "coauthor" name list field containing the data
%    from the loop variable
% 6. Remove the temporary author field copy "coauthortext"
%
% The special $MAPLOOP variable holds each value of the csv field name by
% the "foreach" option on the \map macro.
% The special $MAPUNIQ variable is a random string generated automatically
% at the beginning of each \map section. This can be when you need a unique
% string as here because entrykeys should be new unique.
\DeclareSourcemap{
  \maps[datatype=bibtex]{
    \map[overwrite]{
       \step[fieldsource=author,
             match=\regexp{Arthur\s+Smith}, final]
       \step[fieldsource=author,
             notmatch=\regexp{^Arthur\s+Smith$}, final]
       \step[fieldsource=author, fieldset=coauthortext, origfieldval]
       \step[fieldsource=coauthortext,
             match=\regexp{Arthur\s+Smith\s+and\s+},
             replace={}]
       \step[fieldsource=coauthortext,
             match=\regexp{\s+and\s+Arthur\s+Smith},
             replace={}]
       \step[fieldsource=coauthortext,
             match=\regexp{\s+and\s+},
             replace={,}]
    }
    \map[overwrite, foreach=coauthortext]{
      \step[fieldsource=coauthortext, match=\regexp{.}, final]
      \step[entrynew=\regexp{$MAPUNIQ}, entrynewtype=coauthor]
      \step[entrytarget=\regexp{$MAPUNIQVAL}, fieldset=coauthor, fieldvalue=\regexp{$MAPLOOP}]
      \step[entrytarget=\regexp{$MAPUNIQVAL}, fieldset=options, fieldvalue=skipbib]
    }
    \map[overwrite]{
       \step[fieldsource=coauthortext, null]
    }
  }
}

% Sort the biblist by coauthor name
\DeclareSortingScheme{coauthor}{
  \sort{
        \field{coauthor}
  }
}

% Driver to print the biblist items
\DeclareBibliographyDriver{coauthor}{%
  \printnames{coauthor}}

% bibcheck to strip duplicates
\defbibcheck{coauthor}{%
  \ifcsdef{\strname{coauthor}}
    {\skipentry}
    {\savenamecs{coauthor}{\strname{coauthor}}}}

% Simple bibenvironment to print the biblist
\defbibenvironment{coauthor}
{\list{}
  {\setlength{\leftmargin}{\bibhang}%
   \setlength{\itemindent}{-\leftmargin}%
   \setlength{\itemsep}{\bibitemsep}%
   \setlength{\parsep}{\bibparsep}}}
{\endlist}
{\item}

\begin{document}
\nocite{*}
\printbibliography
% This will automatically use the "coauthor" sorting scheme/biblist/driver etc.
\printbiblist[title=CoAuthors]{coauthor}
\end{document}

I think I found an answer, using an author index. I've tried to include a MWE below, but I might have added a few lines which are unnecessary.

\documentclass{article}
\usepackage[indexing=true,maxnames=99,style=authoryear,datamodel=\jobname]{biblatex}
\usepackage{filecontents}
\usepackage{imakeidx}

\makeindex[program=makeindex,options=-s mystyle.ist,title=List of Co-Authors]
\usepackage{letltxmacro}

% Declare the coauthor name field
\begin{filecontents}{\jobname.dbx}
\DeclareDatamodelFields[type=list, datatype=name]{coauthor}
\DeclareDatamodelEntryfields{coauthor}
\end{filecontents}

% Required to remove the page number after the author in the index
\makeatletter
\let\mygobble\@gobble
\LetLtxMacro\OldIndex\index
\renewcommand{\index}[1]{\OldIndex{#1|mygobble}} 
\makeatother

\begin{filecontents*}{mystyle.ist}
quote '+'
delim_0 " "
delim_1 " "
delim_2 " "
delim_n " "
\end{filecontents*}

% I use a slightly more general bib, with a single author publication, and co-authors over several papers.
\begin{filecontents}{\jobname.bib}
@ARTICLE{test1,
 AUTHOR = {Arthur Smith and James Jones and Bill Smiley and Jake Wrath},
 JOURNALTITLE = {A Journal},
 YEAR   = {2010}
}

@ARTICLE{test2,
 AUTHOR = {Philippa Smyth and Arthur Smith and Bill Likely},
 JOURNALTITLE = {Another Journal},
 YEAR   = {2011}
}

@ARTICLE{test3,
 AUTHOR = {Arthur Smith},
 JOURNALTITLE = {The Best Journal},
 YEAR   = {2013}
}

@ARTICLE{test4,
 AUTHOR = {Philippa Smyth and Arthur Smith and James Jones},
 JOURNALTITLE = {Again Another Journal},
 YEAR   = {2014}
}
\end{filecontents}

\addbibresource{\jobname.bib}

% Create a "coauthor" name list which is a copy of the "author" name
% list, minus oneself (assuming you are "Arthur Smith" and this is how
% your name always appears in the .bib - you can always make more
% sophisticated regexps if there is name variation)
\DeclareSourcemap{
  \maps[datatype=bibtex]{
    \map[overwrite]{
       \step[fieldsource=author,
             match=\regexp{Arthur\s+Smith}, final]
       \step[fieldsource=author, fieldset=coauthor, origfieldval]
       \step[fieldsource=coauthor,
             match=\regexp{Arthur\s+Smith\s+and\s+},
             replace={}]
       \step[fieldsource=coauthor,
             match=\regexp{\s+and\s+Arthur\s+Smith},
             replace={}]
       \step[fieldsource=coauthor,
             match=\regexp{Arthur\s+Smith},
             replace={}]             
    }
  }
}

% Sort the biblist by coauthor name
\DeclareSortingScheme{coauthor}{\sort{\field{coauthor}}}

\DeclareIndexNameFormat{default}{\usebibmacro{index:name}{\index}{#1}{#3}{#5}{#7}}

\renewbibmacro*{bibindex}{\ifbibindex{\indexnames{coauthor}}{}}


% Driver to print the biblist items
\DeclareBibliographyDriver{coauthor}{\printnames{coauthor}}

% Simple bibenvironment to print the biblist
\defbibenvironment{coauthor}{\list{}{}}{\endlist}{\item}

\begin{document}
\nocite{*}
\printbibliography
\raggedright
\printindex
\end{document}