update blocksparse to use additive attention, more consistent with core

[dianaml0]

What does this PR do?

Update blocksparse to use additive attention, this would be more consistent with the core attention masking, which is now only additive or boolean. Perhaps we can allow either options to be specified?

Before submitting

• Did you have fun?
  • Make sure you had fun coding 🙃
• Did you read the contributor guideline?
• Was this discussed/approved via a Github issue? (no need for typos, doc improvements)
  • N/A
• Did you make sure to update the docs?
  • N/A
• Did you write any new necessary tests?
  • N/A
• Did you update the changelog? (if needed)
  • N/A

PR review

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.

[codecov-commenter]

Codecov Report

Merging #85 (8f3c01e) into main (ec26827) will not change coverage.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main      #85   +/-   ##
=======================================
  Coverage   87.08%   87.08%           
=======================================
  Files          50       50           
  Lines        2447     2447           
=======================================
  Hits         2131     2131           
  Misses        316      316           

Flag	Coverage Δ
Python	87.08% <100.00%> (ø)

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

Impacted Files	Coverage Δ
xformers/components/attention/blocksparse.py	90.27% <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 ec26827...8f3c01e. Read the comment docs.

[blefaudeux]

Nice, thank you @dianaml0 !