Ivan and Alexander write lists of integers. Ivan writes all the lists of length $n$ with elements $a_1,a_2,\dots,a_n$ such that $|a_1| + |a_2|+\dots+|a_n| \le k$. Alexander writes all the lists with length $k$ with elements $b_1,b_2,\dots,b_k$ such that $|b_1|+|b_2|+···+|b_k| \le n$. Prove that Alexander and Ivan wrote the same number of lists.
I was having trouble understanding the proof. I get how there are $\binom{k}{r}$ possible. But I don't get how there are $\binom{n}{r} 2^r$ ways, we have $r+1$ integers summing up to $k+1$. Also, I am not sure why total number of lists are $$\sum_{r=0}\binom{n}{r}\binom{k}{r}2^r.$$
We are introducing $c_0$ which is a positive element. So can someone explain the "bijective" way. As in, if we have elements $c_0,\dots,c_r$ summing to $k+1$. How do we get lists Ivan writes?
I think the author made a mistake as they are counting lists for Ivan and not Alexander.
Please help. Any new solution is also appreciated.

I'm going to assume that you do indeed, as you said, understand that there are $\binom{k}{r}$ ways to have $r$ positive integers with a sum $\le k$.
Remember that Alexander's list has $n$ elements, so the remaining $n-r$ elements in his list must all be zero's. Thus we have to distribute the $r$ positive integers among the $n$ elements in his list, and this can be done in $\binom{n}{r}$ ways.
Also, since the given restriction has absolute-values on each element, each of the $r$ positive integers could be replaced with its opposite, and this can be done in $2^r$ ways.
Thus the number of lists Alexander could write for a fixed value $r$ of non-zero entries is
$$\binom{k}{r} \binom{n}{r} 2^r$$
We have to sum this over all possible values of $r$; the minimum value for $r$ should be $1$, and $r$ cannot go over min$(k,n)$, so that would give us
$$\sum_{r=1}^{\min(k,n)} \binom{k}{r} \binom{n}{r} 2^r$$
If we use the same approach to calculate the number of lists Ivan can write, note the only difference would be the two binomials would be reversed, i.e.
$$\sum_{r=1}^{\min(k,n)} \binom{n}{r} \binom{k}{r} 2^r$$
so the total number of lists each can write is clearly the same.
In the given answer they extend the sum(s) to infinity by adopting the usual convention that $\binom{p}{q} = 0$ whenever $q>p$; I'm not sure why they did that, it's not wrong but it seems unnecessary. They're also starting the sums at $r=0$ which does seem wrong as there cannot be $0$ non-zero terms in either list, but since this simply adds $1$ to each of the two sums it still does not stop them from being equal.