Instead of measuring the snail's (or snake's?) progress in meters, measure it in fractions of the entire rubber band, since that is a measure that stays the same during the nightly stretching.
Then on day $n$ it makes a progress of $1/10n$ rubber bands. So you want find out whether there's a $k$ such that $$ \sum_{n=1}^{k} \frac1{10n} \ge 1 $$ and we know this to be the case since the harmonic series diverges.
* * *
**A more vivid argument:** Start by painting 20 dots on the rubber band at equal intervals. The rubber _between_ two dots stretches by 50 cm every night, but since the snail crawls at 100 cm a day, each morning it will be at least 50 cm closer to the _next_ dot than it was the previous morning (unless it passed a dot yesterday). So it _will_ inevitably reach the next dot. And all it needs to do to get all the way is to "reach the next dot" 20 times in a row.
_(Any fencepost errors in this argument are left for the reader to think around.)_