Parallelized prlist function

[zqumei0]

Issue #, if available:
N/A

Description of changes:
Parallelized for loop in prlist() responsible for making awscli calls and formatting pr lines decreasing runtime of prlist operation.

Runtime measurement below were measure as follows:

START=`date +%s`
# Switch case statement
END=`date +%s`
runtime=$(echo "$END - $START" | bc -l)
echo "Runtime: $runtime seconds"

Runtime test output for sequential prlist():

PR #       Title                                                        Branch                                                                                                                              
12         PR Branch 5                                                  bug/broken-print <- test_branch5 (qumei-Isengard)                                                                                   
11         PR 4                                                         test_branch3 <- test_branch4 (qumei-Isengard)                                                                                       
10         PR 3                                                         feature/cpp-hello <- test_branch3 (qumei-Isengard)                                                                                  
9          Test PR 1                                                    main <- test-branch1 (qumei-Isengard)                                                                                               
8          Feature test 2                                               main <- test_branch2 (qumei-Isengard)                                                                                               
6          Feature Request - Cpp Hello                                  main <- feature/cpp-hello (qumei-Isengard)
5          Bug Fix - Broken Print                                       main <- bug/broken-print (qumei-Isengard)
4          Feature Request - Python Hello                               main <- feature/python-hello (qumei-Isengard)
Runtime: 13 seconds

Runtime test output for parallel prlist():

PR #       Title                                                        Branch                                                                                               
12         PR Branch 5                                                  bug/broken-print <- test_branch5 (qumei-Isengard)
9          Test PR 1                                                    main <- test-branch1 (qumei-Isengard)
11         PR 4                                                         test_branch3 <- test_branch4 (qumei-Isengard)
8          Feature test 2                                               main <- test_branch2 (qumei-Isengard)
4          Feature Request - Python Hello                               main <- feature/python-hello (qumei-Isengard)
6          Feature Request - Cpp Hello                                  main <- feature/cpp-hello (qumei-Isengard)
5          Bug Fix - Broken Print                                       main <- bug/broken-print (qumei-Isengard)
10         PR 3                                                         feature/cpp-hello <- test_branch3 (qumei-Isengard)
Runtime: 5 seconds

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[rkeene]

Thanks @zqumei0 this is much faster !

One thing that I've noticed is that this can make too many queries per second to the AWS CodeCommit endpoint if you have a lot of open PRs.

What might be useful here is to limit the number of inflight API calls.

This can be done by collecting the background process IDs ("$!" after ) &) and once the list of those grows beyond a certain number, wait, clear the list, and continue.