[jdbc.row] Reuse (:cols builder) during fetch-all-columns!

[alexander-yakushev]

Another easy way to get -15% execution time during selects. The boost comes from:

• Looking up (:cols builder) once per row, not per each cell
• identical? instead of = to compare against ::not-found.
• As an extra bonus, counting the column vector directly is slightly faster than protocol-based (next.jdbc.rs/column-count builder).

Select 10k rows benchmark against current master:

Before (master):
Time per call: 4.35 ms   Alloc per call: 8,446,823b   Iterations: 2315
Time per call: 4.33 ms   Alloc per call: 8,446,793b   Iterations: 2371
Time per call: 4.32 ms   Alloc per call: 8,446,796b   Iterations: 2321

After (this PR):
Time per call: 3.67 ms   Alloc per call: 8,446,794b   Iterations: 2761
Time per call: 3.69 ms   Alloc per call: 8,446,795b   Iterations: 2745
Time per call: 3.68 ms   Alloc per call: 8,446,793b   Iterations: 2731

[codecov]

Codecov Report

Attention: Patch coverage is 91.66667% with 1 line in your changes missing coverage. Please review.

Project coverage is 83.56%. Comparing base (edee4e0) to head (9131e5a).

Files with missing lines	Patch %	Lines
src/toucan2/jdbc/row.clj	91.66%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #194      +/-   ##
==========================================
- Coverage   83.60%   83.56%   -0.04%     
==========================================
  Files          37       37              
  Lines        2501     2501              
  Branches      212      212              
==========================================
- Hits         2091     2090       -1     
- Misses        198      199       +1     
  Partials      212      212              

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