tab_model: show.std support for all options given by effectsize?

Currently the show.std arg in tab_model allows only for standardization of parameters via re-fitting. I suggest using and supporting the options in effectsize:

When interaction are involved, the can cause the standardized coefs to not corespond to the same simple slopes as the raw coef:

m <- lm(mpg ~ cyl * hp, data = mtcars)

sjPlot::tab_model(m, show.std = TRUE)

	mpg
Predictors	Estimates	std. Beta	CI	standardized CI	p	std. p
(Intercept)	50.75	-0.32	37.41 – 64.09	-0.67 – 0.02	<0.001	0.065
cyl	-4.12	-0.36	-6.14 – -2.09	-0.79 – 0.07	<0.001	0.097
hp	-0.17	-0.55	-0.31 – -0.03	-1.00 – -0.10	0.020	0.018
cyl \* hp	0.02	0.40	0.00 – 0.04	0.03 – 0.77	0.033	0.033
Observations	32
R² / R² adjusted	0.780 / 0.757

effectsize::standardize_parameters(m, method = "refit")
#> # Standardization method: refit
#> 
#> Parameter   | Coefficient (std.) |         95% CI
#> -------------------------------------------------
#> (Intercept) |              -0.32 | [-0.67,  0.02]
#> cyl         |              -0.36 | [-0.79,  0.07]
#> hp          |              -0.55 | [-1.00, -0.10]
#> cyl:hp      |               0.40 | [ 0.03,  0.77]

# These DO correspond to the same simple slopes!
effectsize::standardize_parameters(m, method = "basic")
#> # Standardization method: basic
#> 
#> Parameter   | Coefficient (std.) |         95% CI
#> -------------------------------------------------
#> (Intercept) |               0.00 | [ 0.00,  0.00]
#> cyl         |              -1.22 | [-1.82, -0.62]
#> hp          |              -1.94 | [-3.55, -0.33]
#> cyl:hp      |               2.21 | [ 0.19,  4.23]

^{Created on 2021-06-20 by the reprex package (v2.0.0)}

For mixed models, it could be nice to support the "pseudo" method (:
I'm sure there are more use cases (:

strengejacke / sjPlot

tab_model: show.std support for all options given by effectsize? #767