Added concurrency using lst-parallel.py

[pradhanhitesh]

Description

The lst-parallel.py script helps execute multiple concurrent lst commands using the CPU and improves processing time for large datasets.

1. The script is designed to take folder paths (--t1, --flair, --output) and the number of concurrent (--max_workers) processes as input to run. All other lst flags can also be added (such as --segment-only, --threshold, --stripped, --clipping, --fast-mode, --probability-map), which will subsequently be added while executing. In this script, I have added --save_temp, a flag that invokes the --temp flag for lst with the appropriate file path to save the temporary files.
2. The script generates three folders that are: Output, Log, and Temp (if --save_temp is set). In the Output folder, subject-wise files are stored. In the Log folder, logs generated by the lst command is added and stored, along with time taken to finish the processing. If save_temp is set, Temp folder stores all the temporary files, subject-wise.
3. The script also compiles the stats for all the subjects and stores in the under name lst_stats.csv in the Output directory.

Related Issue

Fixes #33

Type of Change

• New feature

Checklist

• I have performed a self-review of my code.
• My changes generate no new warnings or errors.
• I have updated the documentation if necessary.
• My code follows the repository's coding standards.
• I have included the relevant labels for this PR.

Additional Notes

Currently, I have not tested --annotate-only and --existing_seg flags in the lst-parallel.py script yet. Therefore, I have excluded them from the script, including --lesion-threshold. To run lst-parallel.py, pandas package is required, which is not supplied initially through pip install -e .