Circular Statistics: astropy vs scipy

Seems like scipy has some circular statistics support that might overlap with astropy a little. However, API is not exactly the same and I am not sure if both sides cover all the same use cases. There is a need to do the following:

1. Compare https://docs.astropy.org/en/latest/stats/circ.html with https://docs.scipy.org/doc/scipy/reference/stats.html#circular-statistical-functions
2. Who do we talk to on the scipy side about this?
3. How do we make them 100% compatible?
4. Have a plan and roadmap to deprecate what we have and point people upstream.
5. If Option 4 is not possible, clearly document why and when user should choose one over the other.

### circmean

astropy: `astropy.stats.circstats.circmean(data[, axis, weights])`
*Computes the circular mean angle of an array of circular data.*

scipy: `scipy.stats.circmean(samples[, high, low, axis, nan_policy])`
*Compute the circular mean for samples in a range.*

### circstd

astropy: `astropy.stats.circstats.circstd(data[, axis, weights, method])`
*Computes the circular standard deviation of an array of circular data.*

scipy: `scipy.stats.circstd(samples[, high, low, axis, ...])`
*Compute the circular standard deviation for samples assumed to be in the range [low to high].*

### circvar

astropy: `astropy.stats.circstats.circvar(data[, axis, weights])`
*Computes the circular variance of an array of circular data.*

scipy: `circvar(samples[, high, low, axis, nan_policy])`
*Compute the circular variance for samples assumed to be in a range.*

Start by comparing Astropy's circstats documentation with SciPy's circular statistical functions for circmean, circstd, and circvar, including their parameters and supported use cases. Identify the appropriate SciPy contact, then produce a compatibility and deprecation roadmap; if upstreaming is not feasible, document the differences and when users should choose each implementation.

Written by the indexing model from the issue text.