use unbuffered channels

[malisetti]

In #126 the output channel does not have to a buffered channel and having buffer length equal to length of total processes is possibly not the best buffer length because there may not be all go related processes. imo, this change makes FindAll function easier to understand in terms of the flow.

[tklauser]

Does this fix an issue in the worker concurreny which #126 didn't address? Why is using unbuffered channels better? Please state in the PR description and/or commit message why this change is necessary, not just what it does.

[malisetti]

@tklauser Updated the description of the PR.

[tklauser]

... and having buffer length equal to length of total processes is possibly not the best buffer length because there may not be all go related processes

I don't see why it would matter that not all of the processes are Go processes. The idea is to account for all currently running processes on the system and then let concurrencyLimit goroutines work to determine whether they are Go processes.

this change makes FindAll function easier to understand in terms of the flow.

I think the current solution is fairly easy to understand and I don't think this PR makes understanding the control flow easier. Unless there is a bug in the current solution, I'd prefer to leave it that way.