Opengl distortion issue in the nudging example of VectorField

Nobody has claimed this yet.

## Description of bug / unexpected behavior

In the nudging example in of VectorField, parts of the circles go in and out of view and gets distorted when using opengl.

Cairo:

https://user-images.githubusercontent.com/32387857/128901100-d0781ad9-32c1-478b-a0c7-051eb329c964.mp4

Opengl:

https://user-images.githubusercontent.com/32387857/128900954-6d789384-bf12-4b92-bf7b-5b1323979967.mp4

## Expected behavior

Expect it to be the same as cairo with no distortion

## How to reproduce the issue

Code for reproducing the problem

```py
class Nudging(Scene):
def construct(self):
func = lambda pos: np.sin(pos[1] / 2) * RIGHT + np.cos(pos[0] / 2) * UP
vector_field = ArrowVectorField(func, delta_x=1, delta_y=1, length_func=lambda x: x / 2)
self.add(vector_field)
circle = Circle(radius=2).shift(LEFT)
self.add(circle.copy().set_color(GRAY))
dot = Dot().move_to(circle)

vector_field.nudge(circle, -2, 60, True)
vector_field.nudge(dot, -2, 60)

circle.add_updater(vector_field.get_nudge_updater(pointwise=True))
dot.add_updater(vector_field.get_nudge_updater())
self.add(circle, dot)
self.wait(6)
```

## Additional media files

Images/GIFs

## Logs
Terminal output

```
PASTE HERE OR PROVIDE LINK TO https://pastebin.com/ OR SIMILAR
```

## System specifications

System Details

- OS (with version, e.g Windows 10 v2004 or macOS 10.15 (Catalina)):
- RAM:
- Python version (`python/py/python3 --version`):
- Installed modules (provide output from `pip list`):
```
PASTE HERE
```

LaTeX details

+ LaTeX distribution (e.g. TeX Live 2020):
+ Installed LaTeX packages:

FFMPEG

Output of `ffmpeg -version`:

```
PASTE HERE
```

## Additional comments

Start with the provided Nudging scene and run it with both OpenGL and Cairo, focusing on ArrowVectorField.nudge and the circle and dot updaters. Compare the rendered motion and trace the OpenGL-specific distortion; done means the OpenGL result matches Cairo without parts of the circle leaving view or becoming distorted.

Written by the indexing model from the issue text.