Enhancement: Plot benchmark results with log axes

[corneliusroemer]

I tried to look at your benchmarks.

Unfortunately, I can see much, because the axes are linear and not logarithmic.

It could be helpful to put at least time on log axis.

Also, tabular results would be great for accessibility and convenience.

[jmonroynieto]

You can learn more about the benchmarks on the paper and the additional data see page 9.

[shenwei356]

Thank you Juan.

I always provide reproducible benchmark scripts and results along with source code. The tabular results are here.

And I keep updating the seqkit and have optimized the speed a lot since the publication which was five years ago. Here's an FASTA/Q I/O benchmark result between seqkit v2.0.0 and the lastet seqtk.

Here' another [FASTQ reading benchmark(https://github.com/lh3/biofast/pull/22) with many languages. Where fqcnt_go2 used my bio library.

app	gzip	plain
fqcnt_needletail	5.924	0.733
fqcnt_c1_kseq	7.402	1.115
fqcnt_rustbio	7.413	2.331
fqcnt_nim1_klib	8.059	1.781
fqcnt_go2	8.987	1.166
fqcnt_go1	23.058	3.534