refactor: Deprecated crnn_resnet31 and sar_vgg16_bn

[fg-mindee]

This PR introduces the following modifications:

• deprecated crnn_resnet31 and sar_vgg16_bn and they use combinations of backbones that aren't recommended by their respective paper and do not bring about perf or speed boost.
• updated documentation accordingly
• updated unittests accordingly
• updated demo reco model selection

Any feedback is welcome!

[charlesmindee]

Thanks!

[codecov]

Codecov Report

Merging #468 (7146b25) into main (8fd05b4) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main     #468      +/-   ##
==========================================
+ Coverage   95.77%   95.79%   +0.01%     
==========================================
  Files          95       95              
  Lines        4000     3992       -8     
==========================================
- Hits         3831     3824       -7     
+ Misses        169      168       -1     

Flag	Coverage Δ
unittests	95.79% <100.00%> (+0.01%)	⬆️

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

Impacted Files	Coverage Δ
doctr/models/recognition/crnn/pytorch.py	98.91% <100.00%> (-0.03%)	⬇️
doctr/models/recognition/crnn/tensorflow.py	100.00% <100.00%> (ø)
doctr/models/recognition/sar/pytorch.py	98.34% <100.00%> (-0.03%)	⬇️
doctr/models/recognition/sar/tensorflow.py	98.49% <100.00%> (-0.03%)	⬇️
doctr/models/recognition/zoo.py	100.00% <100.00%> (ø)
...dels/detection/differentiable_binarization/base.py	91.97% <0.00%> (+0.61%)	⬆️

---

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 8fd05b4...7146b25. Read the comment docs.

[charlesmindee]

Thanks