Optimize PR test workflow (Simple starter version)

[qwang98]

This PR increases the number of test_slow PR test bins from 8 to 17, thereby strictly reducing the time spent on test_slow.

We have three types of runners:

• warp-ubuntu-2404-x64-8x for: build, run_examples, bench. I think this is a paid server: https://www.warpbuild.com/, which @leonardoalt last changed in update CI ubuntu #2187, so I have a question: why don't we use the GitHub Workflow free ones? Besides, I think we can parallelize run_examples, which is a bottleneck of 38 minutes currently run single threaded, so there can be some immediate time improvement if we run it on multiple runner instances. Is this server charged by the number of runners or the number of minutes?
• ubuntu-24.04 for: test_quick, test_estark_polygon, test_slow.
• ubuntu-22.04 for: udeps.

BEFORE Optimization

AFTER Optimization
See run time of this PR.

Future Optimization
In #2580, I'm working on creating 14 bins with 4 threads each because we have 55 tests in total and each can run on a separate thread. This would reduce the test_slow run time to that of the longest test, which is ~10 minutes.

This requires computing the bin-thread assignment because nextest only supports the hash method for partitions, which isn't ideal in this case.

[leonardoalt]

so I have a question: why don't we use the GitHub Workflow free ones?

@qwang98 they're too slow and have too little RAM

[leonardoalt]

Is this server charged by the number of runners or the number of minutes?

Number of computation minutes, so it shouldn't matter how many runners we're using if we're faster/same.

[leonardoalt]

I think we can parallelize run_examples, which is a bottleneck of 38 minutes currently run single threaded

Yea those are bad in general, because they don't re-use the build binaries. I think I/we need to re-think those tests anyway.

[qwang98]

I think we can parallelize run_examples, which is a bottleneck of 38 minutes currently run single threaded

Yea those are bad in general, because they don't re-use the build binaries. I think I/we need to re-think those tests anyway.

I mean we can just create multiple bins for them to run, because we currently run 5 tests single threaded. (Unless our subscription doesn't support multiple concurrent runners).

[qwang98]

I think we can parallelize run_examples, which is a bottleneck of 38 minutes currently run single threaded

Yea those are bad in general, because they don't re-use the build binaries. I think I/we need to re-think those tests anyway.

It's also not a bottleneck yet, because even with this PR, our bottleneck is build + test_slow (the latter depends on the former)...

[qwang98]

so I have a question: why don't we use the GitHub Workflow free ones?

@qwang98 they're too slow and have too little RAM

Got it

[qwang98]

I tested a few parameters and think 17 bins for test_slow is the best so far.

I think this can be merged for some immediate speedups.