Add init_cfg for dense heads.

[liuyanyi]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Add init_cfg for dense heads to prevent multiple initialization.
switch back to orginal pycocotools, pycocotools-windows didn't support python 3.9.

Modification

Add init_cfg for dense heads.
switch back to original pycocotools on Windows

BC-breaking (Optional)

Does the modification introduce changes that break the back-compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
3. The documentation has been modified accordingly, like docstring or example tutorials.

[zytx121]

Pretty works! Thank you so much!
LGTM

[codecov-commenter]

Codecov Report

Merging #37 (2f54e87) into main (ecfdd39) will increase coverage by 0.09%.
The diff coverage is 38.46%.

@@            Coverage Diff             @@
##             main      #37      +/-   ##
==========================================
+ Coverage   28.91%   29.01%   +0.09%     
==========================================
  Files          97       97              
  Lines        6477     6442      -35     
  Branches      970      964       -6     
==========================================
- Hits         1873     1869       -4     
+ Misses       4540     4509      -31     
  Partials       64       64              

Flag	Coverage Δ
unittests	28.99% <38.46%> (+0.09%)	⬆️

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

Impacted Files	Coverage Δ
mmrotate/datasets/dota.py	11.17% <ø> (ø)
...ate/models/dense_heads/kfiou_rotate_retina_head.py	30.00% <ø> (ø)
...els/dense_heads/kfiou_rotate_retina_refine_head.py	25.00% <ø> (ø)
mmrotate/models/dense_heads/rotated_anchor_head.py	10.12% <0.00%> (-0.64%)	⬇️
...e/models/dense_heads/rotated_retina_refine_head.py	25.00% <ø> (ø)
mmrotate/models/detectors/rotated_reppoints.py	83.33% <ø> (+11.90%)	⬆️
mmrotate/models/detectors/rotated_retinanet.py	83.33% <0.00%> (+11.90%)	⬆️
mmrotate/models/detectors/s2anet.py	17.64% <ø> (+0.50%)	⬆️
mmrotate/models/detectors/r3det.py	17.28% <25.00%> (+1.43%)	⬆️
...rotate/models/dense_heads/kfiou_odm_refine_head.py	25.80% <100.00%> (+1.86%)	⬆️
... and 2 more

---

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 7c9a592...2f54e87. Read the comment docs.