feat: Add Retinanet and backbones for detection

[kaushikb11]

What does this PR do?

Add Retinanet & backbones for detection with fpn support.

Support for torchvision and bolts backbones.

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests? [not needed for typos/docs]
• Did you verify new and existing tests pass locally with your changes?
• If you made a notable change (that affects users), did you update the CHANGELOG?

PR review

• Is this pull request ready for review? (if not, please submit in draft mode)

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #121 (25500b3) into master (be07c10) will increase coverage by 0.26%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #121      +/-   ##
==========================================
+ Coverage   87.12%   87.39%   +0.26%     
==========================================
  Files          49       49              
  Lines        1554     1579      +25     
==========================================
+ Hits         1354     1380      +26     
+ Misses        200      199       -1     

Flag	Coverage Δ
unittests	87.39% <100.00%> (+0.26%)	⬆️

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

Impacted Files	Coverage Δ
flash/vision/backbones.py	93.54% <100.00%> (+2.63%)	⬆️
flash/vision/classification/model.py	100.00% <100.00%> (ø)
flash/vision/detection/model.py	73.97% <100.00%> (+8.51%)	⬆️
flash/vision/embedding/image_embedder_model.py	79.66% <100.00%> (ø)

---

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 be07c10...76fa0b0. Read the comment docs.

[tchaton]

This logic could/should be within FinetuningCallback.

If the user requires model= fasterrcnn, then it should choose the FasterRCNNFinetuning Callback.

[kaushikb11]

Both the models will have the same FineTuningCallback as they would have similar backbones, but different heads. But yes, could think of moving trainable_backbone_layers for FineTuningCallback OR we could offer some options for finetuning functionalities to the User and it would override the trainable_backbone_layers.