Unroll some @eval loops around _arb_set code

[lgoettgens]

Ass suggested in #1909 (comment).

This contains #1921 as well to avoid conflicts

[codecov]

Codecov Report

Attention: Patch coverage is 80.32787% with 48 lines in your changes missing coverage. Please review.

Project coverage is 87.42%. Comparing base (eea3f94) to head (cfe8054).
Report is 2 commits behind head on master.

Files with missing lines	Patch %	Lines
src/arb/Complex.jl	73.68%	20 Missing ⚠️
src/arb/acb.jl	73.68%	20 Missing ⚠️
src/arb/Real.jl	91.30%	4 Missing ⚠️
src/arb/arb.jl	91.30%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1922      +/-   ##
==========================================
+ Coverage   87.41%   87.42%   +0.01%     
==========================================
  Files          97       97              
  Lines       35626    35642      +16     
==========================================
+ Hits        31142    31161      +19     
+ Misses       4484     4481       -3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[fingolfin]

If the third attempt to re-run OscarCI also fails maybe we need to look closer at the cause?

[benlorenz]

If the third attempt to re-run OscarCI also fails maybe we need to look closer at the cause?

I had a quick look at the logs, both times it was stuck during vinberg_2 in the booktest (for 4 hours). I have seen similar hangs on the oscar repo as well, but not very often.