Refactor load_from_yolov5 in YOLOv5

[zhiqwang]

Adopt classmethod to call load_from_yolov5 in YOLOv5.

---

NOTE: This PR is bc-breaking, we rename the YOLOModule to YOLOv5, in detail we're moving the following

https://github.com/zhiqwang/yolov5-rt-stack/blob/750e3b4b6649e64f9c7ee06d7aa04abccc1c4124/yolort/models/yolo_module.py#L24-L27

to

https://github.com/zhiqwang/yolov5-rt-stack/blob/845d1b2b8ef7805492c7364ebdd72952efd89b4a/yolort/models/yolo_module.py#L22-L25

[codecov]

Codecov Report

Merging #179 (b301fb3) into master (750e3b4) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #179   +/-   ##
=======================================
  Coverage   96.16%   96.17%           
=======================================
  Files          10       10           
  Lines         600      601    +1     
=======================================
+ Hits          577      578    +1     
  Misses         23       23           

Flag	Coverage Δ
unittests	96.17% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
test/test_models.py	95.39% <100.00%> (+0.02%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 750e3b4...b301fb3. Read the comment docs.