Fermi Dirac fix

[DanielHolleufer]

Made a change such that FermiDiracIntegral(j, x) now agrees with https://npplus.readthedocs.io/en/latest/fermi.html and https://github.com/scott-maddox/fdint when j = -1/2 and x > 2.

Also improved performance. Comparison using BenchmarkTools.jl
Before:

@benchmark FewSpecialFunctions.FermiDiracIntegral(-1 / 2, 1.2)
BenchmarkTools.Trial: 10000 samples with 961 evaluations.
 Range (min … max):   87.409 ns …  27.228 μs  ┊ GC (min … max):  0.00% … 99.21%
 Time  (median):     104.162 ns               ┊ GC (median):     0.00%
 Time  (mean ± σ):   148.603 ns ± 311.127 ns  ┊ GC (mean ± σ):  10.69% ±  9.14%

After:

@benchmark FewSpecialFunctions.FermiDiracIntegral(-1 / 2, 1.2)
BenchmarkTools.Trial: 10000 samples with 1000 evaluations.
 Range (min … max):  0.900 ns … 7.800 ns  ┊ GC (min … max): 0.00% … 0.00%
 Time  (median):     1.000 ns             ┊ GC (median):    0.00%
 Time  (mean ± σ):   1.047 ns ± 0.314 ns  ┊ GC (mean ± σ):  0.00% ± 0.00%

  ▅    █     ▃     ▁                                 ▃    ▁ ▁
  █▁▁▁▁█▁▁▁▁▁█▁▁▁▁▁█▁▁▁▁▇▁▁▁▁▁▆▁▁▁▁▁▃▁▁▁▁▅▁▁▁▁▁█▁▁▁▁▁█▁▁▁▁█ █
  0.9 ns      Histogram: log(frequency) by time      1.9 ns <

 Memory estimate: 0 bytes, allocs estimate: 0.

Also changed Manifest.toml to a newer version and changed support from Documter.jl v0.x.x to v1.x.x.

[MartinMikkelsen]

Really nice changes, thank you for your contribution!

[MartinMikkelsen]

Nice

[MartinMikkelsen]

Good changes, looks good