Feature: Add FastAP Loss

[KarahanS]

• Implementation of FastAP loss. Primarily inspired by Pytorch-metric-learning implementation and official implementation. For more details, refer to the paper https://cs-people.bu.edu/fcakir/papers/fastap_cvpr2019.pdf.
• Additionally, I updated the constructor of Circle Loss to give distance_metric_name to the constructor of the superclass. (I'm not sure if it was not given on purpose or a small mistake - related PR)
• Additionally, updated get_anchor_negative_mask and get_anchor_positive_mask to accept labels_b as Optional[Tensor] so that we can call those functions with only one parameter (as I did in FastAP Loss).
• I added official implementation to the test_fast_ap_loss.py so that we can compare out loss with the "expected" loss value (Inspired by the test file written for pytorch-metric-learning.)

This PR is now open for review. FYI @generall @monatis.
I'd be happy to hear any suggestions, corrections, or improvements.

[netlify]

✅ Deploy Preview for capable-unicorn-d5e336 ready!

Name	Link
🔨 Latest commit	e08d335
🔍 Latest deploy log	https://app.netlify.com/sites/capable-unicorn-d5e336/deploys/64219ab2377bc10008e9ad0b
😎 Deploy Preview	https://deploy-preview-199--capable-unicorn-d5e336.netlify.app/_modules/quaterion/loss/fast_ap_loss
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site settings.

[monatis]

Could you please add a test for this loss as well? You can also run an experiment by using it one of the examples in the repository.

There are two minor issues in the implementation that I commented, otherwise it looks promising.

[monatis]

Thanks for handling these reviews. LGTM, I'll merge it.