R: Passing a list of model objects to broom::glance with purrr produces unexpected errors

Question

I am trying to extract information using broom::glance() from survey models. While I can extract coefficients and other model results from each object using broom:tidy(), I sometimes receive an error that the object I am passing does not exist when I try to obtain information via broom::glance(). I am trying to understand how to address the errors in using glance().

Specifically, it looks like any time I pass a list of survey objects to lapply() or purrr::map() and call broom::glance() within those functions - especially when piping the lists - I attain an error akin to ""Error in .svycheck(design) : object '.' not found".

In the code below, I reproduce the error repeatedly and try to pinpoint the source of the issue. It is difficult to tell if this is a consequence of purrr, survey, or broom having unexpected behavior. What I am expecting is glance to return an error that the model type I passed to it is not supported or the model information it usually provides.

# Packages
library(dplyr)
library(purrr)
library(survey)
library(broom)

# Create example list of data.frames
df_list <- data.frame(z = rnorm(20),
                      y = rnorm(20)) %>%
  list() %>%
  rep(2)

# Turn data.frames into design objects
objs_survey <- df_list %>%
  purrr::map(~ survey::svydesign(
    data = .x, ids = ~ 1, weights = ~ 1
  ))

# Fit models
mods_survey <- objs_survey %>%
  purrr::map(~ survey::svyglm(formula = y ~ z, design = .))

# Works with tidy
mods_survey %>% purrr::map( ~ broom::tidy(.))

# But doesn't work with glance
# "Error in .svycheck(design) : object '.' not found"
mods_survey %>% purrr::map( ~ broom::glance(.))

# Since it looks like the issue is
# the design argument is looking for
# .x and this may be a purrr issue
# let's avoid purrr

# First try manually on a single data.frame

# 1 df: Assigning design without %>% - WORKS
survey::svyglm(formula = y ~ z,
               design = objs_survey[[1]]) %>%
  broom::glance()

# 1d df: Assigning design with %>% - same error
objs_survey[[1]] %>% survey::svyglm(formula = y ~ z,
                                    design = .) %>%
  broom::glance()

# Also issues with lapply - same error
lapply(objs_survey, function(.) {
  survey::svyglm(formula = y ~ z,
                 design = .) %>%
    broom::glance(.)
})