New findmin/max implementation using single-pass reduction

[maleadt]

This PR replaces the two-kernel findminmax implementation to a single-pass one by @tkf, #484 (comment), as proposed by @Ellipse0934, #320 (comment). I only rebased it and cleaned-up a little.

Performance is about as @Ellipse0934 reported, i.e. massive improvements in some cases, and a small regression in others (which was much less pronounced with my GPU). As the implementation also fixes a bug, I'd say that's worth it. FWIW, I don't think the performance penalty comes from not using warp intrinsics, but due to the bloaty code generated by a tuple-heavy reduce closure. And recent/beefy GPUs like mine are often less sensitive to that.

Fixes #553, fixes #320 (comment)

[codecov]

Codecov Report

Merging #576 (042ab92) into master (5f6231d) will decrease coverage by 0.08%.
The diff coverage is 94.44%.

@@            Coverage Diff             @@
##           master     #576      +/-   ##
==========================================
- Coverage   80.20%   80.11%   -0.09%     
==========================================
  Files         116      116              
  Lines        6900     6875      -25     
==========================================
- Hits         5534     5508      -26     
- Misses       1366     1367       +1     

Impacted Files	Coverage Δ
src/indexing.jl	98.18% <94.44%> (-1.82%)	⬇️

---

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 5f6231d...042ab92. Read the comment docs.