Fix n and length of rate vector in regimens joined by t_dose_update

[karawoo]

When joining regimens using t_dose_update, only the rate values from the first regimen were included, and the value in n was incorrect. This fixes both.

[jasmineirx]

should we also update the $rate and $n fields for all the other joins in this function?

[karawoo]

I thought this was already working for the other branches of the logic because the following works:

library("PKPDsim")
reg1 <- new_regimen(
  amt = 1,
  times = c(0, 24),
  type = "oral"
)
reg2 <- new_regimen(
  amt = 2,
  time = c(0, 24, 48),
  type = "oral"
)
reg_b <- join_regimen(reg1, reg2, dose_update = 2)
str(reg_b)
#> List of 8
#>  $ interval       : num 24
#>  $ n              : int 4
#>  $ type           : chr [1:4] "oral" "oral" "oral" "oral"
#>  $ t_inf          : num [1:4] 0 0 0 0
#>  $ dose_times     : num [1:4] 0 24 48 72
#>  $ dose_amts      : num [1:4] 1 2 2 2
#>  $ rate           : num [1:4] 0 0 0 0
#>  $ first_dose_time: POSIXct[1:1], format: "2024-01-08 18:05:03"
#>  - attr(*, "class")= chr [1:2] "regimen" "list"

^{Created on 2024-01-08 with reprex v2.0.2}

But actually, it only works by filling in the default rate of 0 because these are oral regimens. However when I tried to update the logic, I realized that new_regimen() actually does not return rate at all for infusion regimens:

library("PKPDsim")
str(new_regimen(amt = 1, times = 0, type = "infusion", rate = 0.5))
#> List of 7
#>  $ interval       : num 24
#>  $ n              : int 1
#>  $ type           : chr "infusion"
#>  $ t_inf          : num 2
#>  $ dose_times     : num 0
#>  $ dose_amts      : num 1
#>  $ first_dose_time: POSIXct[1:1], format: "2024-01-08 18:04:32"
#>  - attr(*, "class")= chr [1:2] "regimen" "list"

^{Created on 2024-01-08 with reprex v2.0.2}

So to fix join_regimen() we'd need to also update new_regimen() so that rate is included for infusions. I can do this, but I wonder if it would instead make more sense to remove rate from non-infusion regimens instead? It seems a bit odd that we are currently including it only in regimens where it is not relevant.

[jasmineirx]

When we simulate using the regimens, if there are non-zero infusion rates we calculate rate in create_event_table and otherwise we set rate to zero

https://github.com/InsightRX/PKPDsim/blob/master/R/create_event_table.R#L120-L131

so it's not clear to me how we are using rate -- possibly in other checks or times we extract info from regimens. I am ok with merging for now and tackling separately making rate work more consistently.

[karawoo]

Sounds good, I added a separate ticket to address the rate issue more generally