This is the hard version of the problem. The difference between the versions is that in this version, (n \leq 5000). You can hack only if you solved all versions of this problem. Legend has it that when Gauss was a young schoolboy, his teacher tasked the class with summing the integers from (1) to (100), likely as a way to keep them occupied for a while. However, young Gauss quickly came up with the formula (\text{sum} = \frac{n(n+1)}{2}) and found the answer in mere moments. Centuries later, Gauss appears before you in a nightmare with a daunting task... You are given a positive integer (n), find a sequence of integers (a_1, a_2, \ldots, a_n) such that (1 \leq a_i \leq 10^{18}) for all (1 \leq i \leq n), and the GCDs of pairwise adjacent elements of (a) are all distinct. Formally, Additionally, (a) should have the minimum possible number of distinct elements. Each test contains multiple test cases. The first line contains the number of test cases (t) ((1 \le t \le 200)). The description of the test cases follows. The first and only line of each test case contains a single integer (n) ((2 \leq n \leq 5000)) — the size of the sequence to be found. For each test case, output (n) space-separated positive integers (a_1, a_2, \ldots, a_n) ((1 \leq a_i \leq 10^{18})) on a new line that satisfy the condition in the statement. If there are multiple solutions, print any of them. It can be proven that under the problem constraints, a solution always exists. For the second test case, the GCDs of adjacent elements are (\gcd(1, 4), \gcd(4, 4), \gcd(4, 6), \gcd(6, 6) = 1, 4, 2, 6), all of which are distinct. For the third test case, the GCDs of adjacent elements are (\gcd(4, 4), \gcd(4, 6), \gcd(6, 6), \gcd(6, 9), \gcd(9, 9), \gcd(9, 4) = 4, 2, 6, 3, 9, 1), all of which are distinct. For each test case, it can be proven that no sequence with fewer distinct elements exists such that all