R data long-wide restructure

Question

I am currently working with data in R. My dataset looks like this (except with about 3 million observations):

This is the current structure of the data:

And I have two objectives...

Objective 1 is to structure it so that it looks like this:

then my second objective is to go back to the original structure and make it look like this:

I have tried variations of aggregate with dcast (which is apparently being deprecated...?)

so, for example, I have tried this:

df2 <- dcast(df1, Store + Sales~ Year, value.var = "Sales")

or even

df %>% 
  group_by(Store, Year) %>%
  summarise(across(starts_with('Sales'), sum))

And I get a diagonal of sales totals across years, but then I'm unable to summarize them so that it looks like

Store Year1 Year2
A       $$   $$
B       $$   $$

Since there is so much data, it looks like a bunch of stacked identity matrices ...except, instead of 1's there are sales values for the years (there are many many years, not just two).

I am looking for suggestions on how to proceed. One package I found was 'pivottabler' and I have not used it yet, I wanted to see if anyone had any better suggestions first.

:::Much appreciated:::