[Windows]-P 128 --bidir iperf server-side CPU usage is too high, resulting in low performance

* Version of iperf3:
iperf 3.16 (cJSON 1.7.15) (windows ver)

* Hardware:
CPU: Intel Xeon Silver 4410Y*2
Memory: 128GB(DDR5 4000Mhz 16GB*32)
HBA: Intel E810 100G

* Operating system (and distribution, if any):
Windows Server 2022

PS:I know there's a note that there won't be Windows support, but since I had the problem I thought I'd raise it here and it's okay if it doesn't work out.

* Other relevant information (for example, non-default compilers,
libraries, cross-compiling, etc.):
https://github.com/ar51an/iperf3-win-builds

_Please fill out one of the "Bug Report" or "Enhancement Request"
sections, as appropriate. Note that submissions of bug fixes, new
features, etc. should be done as a pull request at
https://github.com/esnet/iperf/pulls_

# Bug Report
* Expected Behavior
Server CPU usage same as Client CPU usage.

* Actual Behavior
Server CPU usage 100%
Client CPU usage 12%
The difference between the two sides is too great, resulting in low performance on the server side.

* Steps to Reproduce
Server cmd> iperf.exe -s
Client cmd> iperf.exe -c IP -P 128 --bidir

* Possible Solution
Optimize the workflow of iperf -s

# Enhancement Request

* Current behavior

* Desired behavior
Maximize the performance of bidirectional testing

* Implementation notes