If we don't limit the sizes of the intermediate results, they will grow
indefinitely, causing each iteration to take longer and longer (in both
memcpy time, and algorithm runtime).
While calculating the trimmed length is fairly expensive, it's a small
cost to pay for uniform iteration times.
(...and ASSERT_NOT_REACHED => VERIFY_NOT_REACHED)
Since all of these checks are done in release builds as well,
let's rename them to VERIFY to prevent confusion, as everyone is
used to assertions being compiled out in release.
We can introduce a new ASSERT macro that is specifically for debug
checks, but I'm doing this wholesale conversion first since we've
accumulated thousands of these already, and it's not immediately
obvious which ones are suitable for ASSERT.
`length` is only the (trimmed) size of the word vector, so we have to
multiply it with the size of each element to ensure all bytes are
compared.
Fixes#5335.
