Add GLM with design-based standard errors

[nadiaenh]

Building on Ayush's code for GLM:

• renamed glm( ) to svyglm( ) to avoid ambiguity and be consistent with R naming
• changed variable names and added comment for clarity
• added test for GLM's in bsrs
• added documentation for GLM's
• improved docstring
• use Survey.variance( ) to calculate SE's after closing New data structures to support multiple replicate methods. #297

Ready to be merged as of August 24, 2023

[ayushpatnaikgit]

1. Add tests for other families also, such as logistic regression.
2. Add documentation. Perhaps a tutorial in https://xkdr.github.io/Survey.jl/dev/getting_started/#Tutorial

Also, we may name it glm instead of svyglm. Just for now, we should avoid the ambiguity.

This PR solves issues #295

[codecov-commenter]

Codecov Report

Merging #304 (c677962) into 0.3.1 (b5b9b09) will decrease coverage by 12.50%.
Report is 6 commits behind head on 0.3.1.
The diff coverage is 92.10%.

@@             Coverage Diff              @@
##             0.3.1     #304       +/-   ##
============================================
- Coverage   100.00%   87.50%   -12.50%     
============================================
  Files           14       15        +1     
  Lines          242      272       +30     
============================================
- Hits           242      238        -4     
- Misses           0       34       +34     

Files Changed	Coverage Δ
src/Survey.jl	100.00% <ø> (ø)
src/quantile.jl	87.50% <57.14%> (-12.50%)	⬇️
src/ratio.jl	75.00% <70.00%> (-25.00%)	⬇️
src/bootstrap.jl	100.00% <100.00%> (ø)
src/by.jl	100.00% <100.00%> (ø)
src/jackknife.jl	100.00% <100.00%> (ø)
src/mean.jl	100.00% <100.00%> (ø)
src/reg.jl	100.00% <100.00%> (ø)
src/total.jl	100.00% <100.00%> (ø)

... and 1 file with indirect coverage changes

[ayushpatnaikgit]

1. Can you try to change the name of the function to glm from svyglm?
2. Can you add a doctest for glm?

[ayushpatnaikgit]

Doctests were passing because :fix = true was uncommented in docs/project.toml this means that all doctests may have incorrect values.

[ayushpatnaikgit]

Bunch of doctests were failing. I am unsure why the doctest CI didn't give an error. I'll fix them in PR 314