Convert daxpy example to alpaka

[bernhardmgruber]

This PR converts the daxpy example to alpaka so it can be used on more architectures.

[codecov]

Codecov Report

Merging #469 (3442c07) into develop (2b903f1) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

❗ Current head 3442c07 differs from pull request most recent head ff18b95. Consider uploading reports for the commit ff18b95 to get more accurate results

@@             Coverage Diff             @@
##           develop     #469      +/-   ##
===========================================
- Coverage    99.24%   99.23%   -0.01%     
===========================================
  Files           59       59              
  Lines         5395     5387       -8     
===========================================
- Hits          5354     5346       -8     
  Misses          41       41              

Impacted Files	Coverage Δ
include/llama/mapping/ChangeType.hpp	100.00% <ø> (ø)
include/llama/Core.hpp	95.23% <100.00%> (-0.32%)	⬇️
include/llama/mapping/Bytesplit.hpp	100.00% <100.00%> (ø)
tests/core.cpp	100.00% <0.00%> (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 2b903f1...ff18b95. Read the comment docs.

[bernhardmgruber]

A quick test on a big Xeon shows:

8192Mi doubles (65536MiB data)
Threads: 80
Affinity: 0-79
baseline std::vector
alloc 75.859 s
init 16.1799 s
daxpy 18.2234 s
daxpy 2.11368 s
daxpy 2.11574 s
daxpy 2.11507 s
daxpy 2.11447 s

We clearly need a warmup run here.

[bernhardmgruber]