Does the data look normal?

You will write two quick checks of a data set: the mean against the median, and the share within $k$ standard deviations. Then you will write a function that decides whether the data look normal. The checker uses two made-up data sets made with a fixed seed.

The idea. For a normal shape, the mean and the median are close, about $68\%$ of the values lie within one standard deviation and about $95\%$ within two. Use the population standard deviation, statistics.pstdev.

Your tasks.

1. Finish share_within(data, k).
2. Finish mean_minus_median(data), measured in standard deviations: $\dfrac{\text{mean} - \text{median}}{\text{sd}}$.
3. Finish looks_normal(data). Return True only when three things hold. The share within $1$ sd is from $0.63$ to $0.73$. The share within $2$ sd is from $0.93$ to $0.97$. The mean minus the median is less than $0.1$ sd either way.