Skip checksum benchmarks on systems with slow cpu

The checksum benchmarking on module load may take a really long time
on embedded systems with a slow cpu. Avoid all benchmarks >= 1MiB on
systems, where EdonR is slower then 300 MiB/s.

This limit is currently hardcoded via the define LIMIT_PERF_MBS.

This is the new benchmark output of a slow Intel Atom:

implementation    1k    4k   16k   64k  256k    1m    4m   16m
edonr-generic    209   257   268   259   262     0     0     0
skein-generic    129   150   151   150   150     0     0     0
sha256-generic    50    55    56    56    56     0     0     0
sha512-generic    76    86    88    89    88     0     0     0
blake3-generic    63    62    62    62    61     0     0     0
blake3-sse2      114   292   301   307   309     0     0     0

Reviewed-by: Sebastian Gottschall <s.gottschall@dd-wrt.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tino Reichardt <milky-zfs@mcmilk.de>
Closes #13695