perf: improve dot distance computing

[BubbleCal]

for 8bit:
time: [582.61 µs 585.08 µs 587.42 µs]
change: [-9.7495% -6.2223% -2.3288%] (p = 0.01 < 0.10)
Performance has improved.

for 4bit:
time: [549.20 µs 551.13 µs 553.55 µs]
change: [-38.117% -37.734% -37.350%] (p = 0.00 < 0.10)
Performance has improved.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 78.04%. Comparing base (9c08d2d) to head (a673bf3).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3169      +/-   ##
==========================================
+ Coverage   78.01%   78.04%   +0.02%     
==========================================
  Files         243      243              
  Lines       82890    82889       -1     
  Branches    82890    82889       -1     
==========================================
+ Hits        64670    64690      +20     
- Misses      14999    15003       +4     
+ Partials     3221     3196      -25     

Flag	Coverage Δ
unittests	78.04% <100.00%> (+0.02%)	⬆️

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

---

🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests

[westonpace]

Nice improvement!