[Bug-gsl] Jacobian elliptic functions with error estimate

[Liam Healy]

I noticed an oddity in the Jacobian elliptic functions function.  It is
named gsl_sf_elljac_e, but despite the "_e" at the end,
it produces no error estimates, just three plain doubles.
Should this be renamed gsl_sf_elljac to be consistent
with what appears to be a naming convention for special
functions?  Also, should there be another function with
that name that does produce error estimates?

Liam