Fix tidy_poisson() p calculation

[spsanderson]

Describe the bug
There is an issue in the calculation of the p column for tidy_poisson()

To Reproduce
Steps to reproduce the behavior:

> TidyDensity::tidy_poisson()
# A tibble: 50 x 7
   sim_number     x     y      dx      dy     p     q
   <fct>      <int> <int>   <dbl>   <dbl> <dbl> <dbl>
 1 1              1     1 -0.921  0.00554 0.368     0
 2 1              2     1 -0.823  0.0138  0.368     0
 3 1              3     2 -0.724  0.0310  0.368     0
 4 1              4     0 -0.625  0.0626  0.368     0
 5 1              5     0 -0.526  0.114   0.368     0
 6 1              6     1 -0.427  0.188   0.368     0
 7 1              7     0 -0.328  0.279   0.368     0
 8 1              8     1 -0.230  0.374   0.368     0
 9 1              9     2 -0.131  0.451   0.368     0
10 1             10     2 -0.0319 0.493   0.368     0
# ... with 40 more rows

Which gives the following probability plot:

Expected behavior

> tidy_poisson()
# A tibble: 50 x 7
   sim_number     x     y     dx      dy     p     q
   <fct>      <int> <int>  <dbl>   <dbl> <dbl> <dbl>
 1 1              1     2 -1.44  0.00429 0.736     0
 2 1              2     0 -1.30  0.00985 0.920     0
 3 1              3     0 -1.16  0.0208  0.981     0
 4 1              4     0 -1.02  0.0403  0.996     0
 5 1              5     0 -0.881 0.0717  0.999     0
 6 1              6     0 -0.741 0.117   1.00      0
 7 1              7     1 -0.600 0.176   1.00      0
 8 1              8     1 -0.460 0.244   1.00      0
 9 1              9     1 -0.319 0.310   1.00      0
10 1             10     0 -0.179 0.365   1.00      0
# ... with 40 more rows

Which gives the following probability plot:

The p line needs to be the following:

dplyr::mutate(p = list(stats::ppois(unlist(x), lambda = lambda))) %>%