fix: revert initial response calculation for lagged adstock in allocator

Current allocator initial response calculation tries to account for only immediate response, which causes unmatched initial response on the saturation curve. Revert back to using total initial spend and thus full initial response, which means the allocator simulates the peak response in lagged adstock.

R/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: Robyn
 Type: Package
 Title: Semi-Automated Marketing Mix Modeling (MMM) from Meta Marketing Science 
-Version: 3.10.5.9001
+Version: 3.10.5.9002
 Authors@R: c(
     person("Gufeng", "Zhou", , "gufeng@meta.com", c("cre","aut")),
     person("Leonel", "Sentana", , "leonelsentana@meta.com", c("aut")),

R/R/allocator.R

@@ -286,11 +286,12 @@ robyn_allocator <- function(robyn_object = NULL,
     names(hist_carryover_temp) <- resp$date[ndates_loc]
     hist_carryover[[i]] <- hist_carryover_temp
     # get simulated response
-    if (resp$input_immediate[1] == initSpendUnit[i]) {
-      x_input <- initSpendUnit[i]
-    } else {
-      x_input <- mean(resp$input_immediate)
-    }
+    # if (resp$input_immediate[1] == initSpendUnit[i]) {
+    #   x_input <- initSpendUnit[i]
+    # } else {
+    #   x_input <- mean(resp$input_immediate)
+    # }
+    x_input <- initSpendUnit[i]
     resp_simulate <- fx_objective(
       x = x_input,
       coeff = coefs_sorted[[mediaSpendSorted[i]]],