RadialVelocity now seems to require a normal vector

[yt-fido]

Originally reported by: Matt Turk (Bitbucket: MatthewTurk, GitHub: MatthewTurk)

---

Radial Velocity now throws an error, because of the nested sph_theta field, when a normal vector is not defined.

I think we could either not use the normal to define the radial velocity (it's not entirely clear to me why we are now) or default to the z-axis.

---

• Bitbucket: https://bitbucket.org/yt_analysis/yt/issue/451

[yt-fido]

Original comment by chummels (Bitbucket: chummels, GitHub: chummels):

---

Removing milestone: 2.5 (automated comment)

[yt-fido]

Original comment by Nathan Goldbaum (Bitbucket: ngoldbaum, GitHub: ngoldbaum):

---

Setting a default value for the 'normal' field parameter. Closes #451