response(<mdl_ts>) is unable to backtransform from parameters existing in the data

Should the model object also store these parameters?
From #103

library(fable)
#> Loading required package: fabletools
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
uad <- as_tsibble(USAccDeaths)
uad %>% 
  bind_cols(features(uad, value, feasts::guerrero)) %>% 
  model(SNAIVE(box_cox(resp(value), lambda_guerrero))) %>% 
  response()
#> Error: Problem with `mutate()` input `response`.
#> x object 'lambda_guerrero' not found
#> ℹ Input `response` is `map(.fit, response)`.

^{Created on 2021-01-08 by the reprex package (v0.3.0)}

Reproduce the reported example using response() with SNAIVE(box_cox(resp(value), lambda_guerrero)) after adding lambda_guerrero via features(). Trace why response() cannot find the existing parameter, then ensure the model object retains enough information for response() to backtransform the data; the example should complete without the object-not-found error.