Finish TinyMS hub module with load and publish feature

[leonwanghui]

What type of PR is this?

/kind enhancement

What does this PR do / why do we need it:

This PR is proposed to add the support of TinyMS model loading and publishing functions.

Model load

For those who are new to TinyMS project, they can leverage tinyms.hub to easily load pretrained model and apply model evaluation and inference operation. Take LeNet5 model for example:

from PIL import Image
from tinyms import hub
from tinyms.vision import mnist_transform
from tinyms.model import Model

img = Image.open(img_path)
img = mnist_transform(img)

# load LeNet5 pretrained model
net= hub.load('tinyms/0.2/lenet5_v1_mnist', class_num=10)
model = Model(net)

res = model.predict(ts.expand_dims(ts.array(img), 0)).asnumpy()
print("The label is:", mnist_transform.postprocess(res))

Model support

Currently we add five model assets into TinyMS Hub:

• tinyms/0.2/lenet5_v1_mnist
• tinyms/0.2/resnet50_v1_cifar10
• tinyms/0.2/resnet50_v1_mushroom
• tinyms/0.2/mobilenet_v2_cifar10
• tinyms/0.2/ssd300_v1_voc2007

Which issue(s) this PR fixes:

Fixes #

Special notes for your reviewers: