When multiplying a uint64 and uint128 and assigning to
a uint256, the solidity checked math applies to the
types of the numbers being multiplied. This means that
solidity catches an overflow when the result of larger than
a uint128 even though its assigned to a uint256.
This PR uses unchecked math to prevent this kind of overflow
instead of casting the values to uint256 because it is
safe to use unchecked math here. The following inequality
always holds true, which shows that it is safe:

``` solidity
type(uint64).max * type(uint128).max < type(uint256).max
```