Eventually remove the bias-correction

[sdtemple]

The parametric bootstrap bias correction functions *_bc() in inference.py are not the best bootstrap methods to address bias correction (in theory and practice). In R, better approaches would be:

• https://jeksterslabds.github.io/jeksterslabRboot/reference/bc.html
• https://jeksterslabds.github.io/jeksterslabRboot/reference/dot-bca.html

In Temple, Waples, and Browning (2024) and Temple (2024), we did not use the bias-corrected intervals. This is because the selection coefficient estimator was already unbiased (empirically) and a function of an unbiased estimator of the IBD rate. We found worse coverage empirically when we did bias correct.

At the next release of this package, if that ever happens, I will remove these two functions. For now, I recommend that you DO NOT use the *_bc() functions.

[sdtemple]

I should also remove the call to bootstrap_standard_bc() in the readthedocs API.