Improve ballista performance

**Is your feature request related to a problem or challenge? Please describe what you are trying to do.**

To be an efficient distributed execution platform, there're still many aspects we need to improve.

**Describe the solution you'd like**

- For scheduler, efficient task scheduling
- [x] #133 #134
- [x] #131 #135
- [x] #137 #147
- [x] #144 #145
- [x] #138 #141
- [x] #149 #150
- [x] #130 #153
- [x] #179 #181
- [x] #136 #156
- [x] #208 #256
- [x] #209 #255
- [x] #391 #392
- [x] #142 #393
- [x] #389 #394
- [x] #371 #395
- [x] #132 #396
- [ ] #397 #398
- [x] https://github.com/apache/arrow-ballista/issues/243

**Describe alternatives you've considered**

**Additional context**

The issue is a broad performance umbrella for Ballista, with efficient task scheduling as the only named area. Review the scheduler-related issues listed in the checklist, especially the unchecked #397 and #398, and trace their current status before starting. There is no file, test, entry point, or concrete completion criterion specified here.

Written by the indexing model from the issue text.