Re-factor mw_neat_output_mfaz() to allow multiple grouping variables

[tomaszaba]

As of now, when users request {mwana} plausibility check neat functions to return outputs grouped by say "province" and then by "district", it does not respond adequately to:

Without colling mw_neat_output_mfaz() it work OK.

> anthro.02 |> group_by(province, strata) |> mw_plausibility_check_mfaz(sex = sex, muac = muac, age = age, flags = flag_mfaz)
# A tibble: 4 × 19
  province strata     n flagged flagged_class sex_ratio sex_ratio_class age_ratio age_ratio_class   dps dps_class      sd sd_class     skew
  <chr>    <chr>  <int>   <dbl> <fct>             <dbl> <chr>               <dbl> <chr>           <dbl> <chr>       <dbl> <chr>       <dbl>
1 Nampula  Rural    472 0.00424 Excellent         0.490 Excellent           0.410 Excellent         100 Problematic 0.895 Good     -0.113  
2 Nampula  Urban    614 0.0130  Good              0.493 Excellent           0.241 Excellent         100 Problematic 0.989 Excelle… -0.364  
3 Zambezia Rural    368 0.0109  Good              0.566 Excellent           0.449 Excellent         100 Problematic 0.988 Excelle…  0.00407
4 Zambezia Urban    813 0.00738 Excellent         1     Excellent           0.391 Excellent         100 Problematic 0.940 Excelle… -0.0430 
# ℹ 5 more variables: skew_class <fct>, kurt <dbl>, kurt_class <fct>, quality_score <dbl>, quality_class <fct>

But once called, the output table gets distorted.

anthro.02 |> group_by(province, strata) |> mw_plausibility_check_mfaz(sex = sex, muac = muac, age = age, flags = flag_mfaz) |> mw_neat_output_mfaz()
# A tibble: 4 × 19
  `Total children` `Flagged data (%)` `Class. of flagged data` `Sex ratio (p)` `Class. of sex ratio` `Age ratio (p)` `Class. of age ratio`
  <chr>            <chr>                                 <int> <chr>           <fct>                 <chr>           <chr>                
1 Nampula          Rural                                   472 0.4%            Excellent             0.490           Excellent            
2 Nampula          Urban                                   614 1.3%            Good                  0.493           Excellent            
3 Zambezia         Rural                                   368 1.1%            Good                  0.566           Excellent            
4 Zambezia         Urban                                   813 0.7%            Excellent             >0.999          Excellent            
# ℹ 12 more variables: `DPS (#)` <chr>, `Class. of DPS` <chr>, `Standard Dev* (#)` <dbl>, `Class. of standard dev` <chr>,
#   `Skewness* (#)` <dbl>, `Class. of skewness` <chr>, `Kurtosis* (#)` <dbl>, `Class. of kurtosis` <fct>, `Overall score` <dbl>,
#   `Overall quality` <fct>, `` <dbl>, `` <dbl>

Here, values under column "strata" get filled under "n" overriding "n"'s.

Task

Fix mw_neat_output_mfaz()