Fixes earth GLM models and prediction test routine

[edgararuiz]

• Fixes issue handling GLM Binomial earth models (Binomial MARS model returns Gaussian coefficients when parsed in tidypredict_fit() and tidypredict_sql() #97)
• Fixes issue on the prediction testing function, it was not working correctly for earth models

[edgararuiz]

Results now match:

options(scipen = 999)
library(tidyverse)
library(tidypredict)
library(earth)
data("etitanic", package = "earth")

mars.mod.1 <- earth(
    survived ~ age + sex,
    data = etitanic,
    glm = list(family = binomial)
  )

etitanic %>%
  tidypredict_to_column(mars.mod.1) %>%
  mutate(pred = predict(mars.mod.1, type = "response")[,1]) %>%
  head(10) 
#>    pclass survived    sex     age sibsp parch       fit      pred
#> 1     1st        1 female 29.0000     0     0 0.7407665 0.7407665
#> 2     1st        1   male  0.9167     1     2 0.4856151 0.4856151
#> 3     1st        0 female  2.0000     1     2 0.9033125 0.9033125
#> 4     1st        0   male 30.0000     1     2 0.1937987 0.1937987
#> 5     1st        0 female 25.0000     1     2 0.7407665 0.7407665
#> 6     1st        1   male 48.0000     0     0 0.1937987 0.1937987
#> 7     1st        1 female 63.0000     1     0 0.7407665 0.7407665
#> 8     1st        0   male 39.0000     0     0 0.1937987 0.1937987
#> 9     1st        1 female 53.0000     2     0 0.7407665 0.7407665
#> 10    1st        0   male 71.0000     0     0 0.1937987 0.1937987

^{Created on 2021-09-28 by the reprex package (v2.0.1)}

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.