Split series over fq_nmod and fq into different files

[lgoettgens]

Inlining makes everything a lot more readable (at least for me). Furthermore, it enables us to grep for some flint function and find all places where it is called (of course if all similar places are inlined/unrolled as well).
And it makes the script in #1812 (comment) a lot better in identifying the flint functions that we call from different locations

The corresponding tests are already in different files.

[codecov]

Codecov Report

Attention: Patch coverage is 92.72727% with 120 lines in your changes missing coverage. Please review.

Project coverage is 87.50%. Comparing base (13e29a5) to head (a4a3ca6).
Report is 15 commits behind head on master.

Files with missing lines	Patch %	Lines
src/flint/fq_nmod_rel_series.jl	92.17%	36 Missing ⚠️
src/flint/fq_rel_series.jl	93.66%	29 Missing ⚠️
src/flint/fq_nmod_abs_series.jl	92.49%	28 Missing ⚠️
src/flint/fq_abs_series.jl	92.47%	27 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1929      +/-   ##
==========================================
+ Coverage   87.39%   87.50%   +0.11%     
==========================================
  Files          97       99       +2     
  Lines       35642    36464     +822     
==========================================
+ Hits        31149    31908     +759     
- Misses       4493     4556      +63     

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

[fingolfin]

I am somewhat wary of increasing code duplication.

On the other hand I understand how it simplifies some tasks.

[lgoettgens]

The long-term goal is of course to reduce the duplication again.
After putting most of the flint calls into small oneline functions, most of the series types should be mostly identical and can then be simplified/deduplicated again.