fix: Fix a couple of buffer overrun hazards.

[bmah888]

Pointed out by @berkakinci.

Fixes #1134.

PLEASE NOTE the following text from the iperf3 license. Submitting a
pull request to the iperf3 repository constitutes "[making]
Enhancements available...publicly":

You are under no obligation whatsoever to provide any bug fixes, patches, or
upgrades to the features, functionality or performance of the source code
("Enhancements") to anyone; however, if you choose to make your Enhancements
available either publicly, or directly to Lawrence Berkeley National
Laboratory, without imposing a separate written license agreement for such
Enhancements, then you hereby grant the following license: a non-exclusive,
royalty-free perpetual license to install, use, modify, prepare derivative
works, incorporate into other computer software, distribute, and sublicense
such enhancements or derivative works thereof, in binary and source code form.

The complete iperf3 license is available in the LICENSE file in the
top directory of the iperf3 source tree.

• Version of iperf3 (or development branch, such as master or
  3.1-STABLE) to which this pull request applies: master

• Issues fixed (if any): Buffer overflow hazard and incorrect return value in iperf_printf(). #1134

• Brief description of code changes (suitable for use as a commit message):
  See above.

[berkakinci]

This part is fixed: now linebuffer[1024] won't overflow.
On setting r and return r:
Under test->role == 'c' There is another couple of missing counts at line 4548 and 4551.
The rest of this is getting in the weeds and would require a bunch of little changes: all of these functions can return -1 on failure, so count wouldn't be correct. Also, the *nprint* functions return the number that would have been written if the buffer was large enough, so count wouldn't be correct if buffers weren't large enough.
If you're not averse to a bigger change: I think both the role == 'c' and role == 's' stanzas can share a lot more code. Set up linebuffer in both cases. Then only take the return value from a final (and single) fprintf() call.

[bmah888]

The attempt to merge this pull request went sideways. Completely my fault. I ended up backing it out and then cherry-picking the individual commits one by one back onto the mainline.