Add the 3d avg pool for video related model (#33339)
Summary:
```
import torch, time
for dtype in [torch.qint8, torch.quint8, torch.qint32]:
print('****', str(dtype), '*****')
x = torch.rand(1, 5, 56, 56, 256)
q_x = torch.quantize_per_tensor(x, 0.5, 1, dtype)
q_x = q_x.permute([0, 4, 1, 2, 3])
x = x.permute([0, 4, 1, 2, 3])
NITER = 10
s = time.time()
for i in range(NITER):
float_out = torch.nn.functional.avg_pool3d(x, kernel_size=3, stride=None, padding=0)
time_per_iter_float = (time.time() - s) / NITER
s = time.time()
for i in range(NITER):
quant_out = torch.nn.quantized.functional.avg_pool3d(q_x, kernel_size=3, stride=None, padding=0)
time_per_iter_quant = (time.time() - s) / NITER
print('time/iter ms (float)', 'time/iter ms (quant)', 'quant/float', sep='\t')
print(time_per_iter_float * 1000, time_per_iter_quant * 1000, time_per_iter_quant / time_per_iter_float, sep='\t')
```
```
**** torch.qint8 *****
time/iter ms (float) time/iter ms (quant) quant/float
16.286182403564453 0.7308721542358398 0.04487682479080417
**** torch.quint8 *****
time/iter ms (float) time/iter ms (quant) quant/float
15.364313125610352 0.6497383117675781 0.042288796541418254
**** torch.qint32 *****
time/iter ms (float) time/iter ms (quant) quant/float
15.649032592773438 13.879132270812988 0.8869003363966556
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/33339
Differential Revision: D19900904
Pulled By: lly-zero-one
fbshipit-source-id: 4522cc6b4a0751aeda6c7edc258e0cb3f55a8fe3
