Put more control in targeting file to parse

[DonaldChung-HK]

Since some pipeline only runs on single system so it might be useful to turn the key to look for file name into a list so that it can target different number and name of log files for different pipelines.

[DonaldChung-HK]

I put in a fix for controlling what file to parse for individual pipeline. it adds the -t for adding list of os keys and -f for adding list of corresponding file names to parse for a pipeline/Job and you use another -t and -f to do this for another one. Here is an example where you only want to monitor Mac in build_packages_from_branch and all in main_nightly_deployment_prototype

python main.py -u 'https://builds.mantidproject.org/' -p 'build_packages_from_branch' 'main_nightly_deployment_prototype' -n 30 -t "Mac" -f "darwin17.log" -t "Mac" "Linux" "Windows" -f "darwin17.log" "linux-gnu.log" "msys.log"

Note that unlike normal command-line application the order which the -t and -f flags are added is important as they correspond to the position of pipeline in -p argument

[DonaldChung-HK]

Probably might require a more robust solution