perf: switch to go-art

[Clement-Jean]

Tried the switch to go-art and here are the results I get in the Put/Get benchmark (which I modified to use Golang benchmark features):

goos: darwin
goarch: arm64
pkg: github.com/ByteStorage/FlyDB/db/engine
cpu: Apple M1 Max
        │   old.txt   │              new.txt               │
        │   sec/op    │   sec/op     vs base               │
Put-10    677.0n ± 1%   656.2n ± 0%  -3.07% (p=0.000 n=50)
Get-10    443.8n ± 1%   441.1n ± 1%       ~ (p=0.222 n=50)
geomean   548.1n        538.0n       -1.84%

        │  old.txt   │               new.txt               │
        │    B/op    │    B/op     vs base                 │
Put-10    526.0 ± 0%   474.0 ± 0%  -9.89% (n=50)
Get-10    304.0 ± 0%   304.0 ± 0%       ~ (p=1.000 n=50) ¹
geomean   399.9        379.6       -5.07%
¹ all samples are equal

        │  old.txt   │               new.txt                │
        │ allocs/op  │ allocs/op   vs base                  │
Put-10    14.00 ± 0%   12.00 ± 0%  -14.29% (n=50)
Get-10    8.000 ± 0%   8.000 ± 0%        ~ (p=1.000 n=50) ¹
geomean   10.58        9.798        -7.42%
¹ all samples are equal

Let me know what you think about this. Also, I can reset the benchmark to the original version if you need to. Just thought it would be a little bit more accurate to use Golang benchmarks.

[github-actions]

Thank you very much for your contribution, we will promptly review your code if there are no errors and pass ci. We will merge your pull request into the master branch.

[qishenonly]

Thanks for your PR! To maintain clean code structure, please reimplement this based on a new data structure rather than modifying the existing one. This helps avoid compatibility issues and simplifies future maintenance. Looking forward to your updates!

[Clement-Jean]

If anyone is willing to implement this the way you want, you can expect even better performance now. Just check the latest go-art commit. On my computer the benchmark looks like this:

goos: darwin
goarch: arm64
pkg: github.com/ByteStorage/FlyDB/db/engine
cpu: Apple M1 Max
        │   old.txt   │              new.txt               │
        │   sec/op    │   sec/op     vs base               │
Put-10    690.2n ± 1%   656.1n ± 3%  -4.94% (p=0.000 n=10)
Get-10    438.3n ± 2%   418.6n ± 3%  -4.48% (p=0.001 n=10)
geomean   550.0n        524.1n       -4.71%

        │  old.txt   │               new.txt                │
        │    B/op    │    B/op     vs base                  │
Put-10    526.0 ± 0%   464.0 ± 0%  -11.79% (p=0.000 n=10)
Get-10    304.0 ± 0%   304.0 ± 0%        ~ (p=1.000 n=10) ¹
geomean   399.9        375.6        -6.08%
¹ all samples are equal

        │  old.txt   │               new.txt                │
        │ allocs/op  │ allocs/op   vs base                  │
Put-10    14.00 ± 0%   12.00 ± 0%  -14.29% (p=0.000 n=10)
Get-10    8.000 ± 0%   8.000 ± 0%        ~ (p=1.000 n=10) ¹
geomean   10.58        9.798        -7.42%