gh-149718: Aggregate same stack frames in Tachyon in some collectors

[maurycy]

Please see gh-149718 for more details.

On a MacBook M4, using test_asyncio since it has longer stacks (without --async-aware):

Sample rate (samples/sec)

Format	main (c6fd7de)	remote-debugger-batching (2864c275847)
--pstats	31,161.59	103,657.01
--collapsed	57,302.45	102,350.22
--flamegraph	33,490.73	100,641.14
--gecko	31,346.37	98,031.88
--heatmap	14,960.64	97,607.28
--jsonl	24,554.56	98,018.93

Error rate (%)

Format	main (c6fd7de)	remote-debugger-batching (2864c275847)
--pstats	1.94	0.54
--collapsed	1.47	0.78
--flamegraph	1.77	0.57
--gecko	1.92	0.57
--heatmap	2.33	0.44
--jsonl	2.06	0.54

Missed samples (%)

Format	main (c6fd7de)	remote-debugger-batching (2864c275847)
--pstats	90.65	68.90
--collapsed	82.81	69.29
--flamegraph	89.95	69.81
--gecko	90.60	70.59
--heatmap	95.51	70.72
--jsonl	92.63	70.59

Bench

mkdir -p /tmp/bench_results

git checkout c6fd7de
sudo -n ./python.exe -m profiling.sampling run -r 300khz --pstats     -o /dev/null            -m test test_asyncio 2>&1 | tee /tmp/bench_results/main_pstats.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --collapsed  -o /dev/null            -m test test_asyncio 2>&1 | tee /tmp/bench_results/main_collapsed.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --flamegraph -o /dev/null            -m test test_asyncio 2>&1 | tee /tmp/bench_results/main_flamegraph.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --gecko      -o /dev/null            -m test test_asyncio 2>&1 | tee /tmp/bench_results/main_gecko.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --heatmap    -o /tmp/heatmap_main    -m test test_asyncio 2>&1 | tee /tmp/bench_results/main_heatmap.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --jsonl      -o /dev/null            -m test test_asyncio 2>&1 | tee /tmp/bench_results/main_jsonl.txt

git checkout remote-debugger-batching
sudo -n ./python.exe -m profiling.sampling run -r 300khz --pstats     -o /dev/null              -m test test_asyncio 2>&1 | tee /tmp/bench_results/batching_pstats.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --collapsed  -o /dev/null              -m test test_asyncio 2>&1 | tee /tmp/bench_results/batching_collapsed.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --flamegraph -o /dev/null              -m test test_asyncio 2>&1 | tee /tmp/bench_results/batching_flamegraph.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --gecko      -o /dev/null              -m test test_asyncio 2>&1 | tee /tmp/bench_results/batching_gecko.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --heatmap    -o /tmp/heatmap_batching  -m test test_asyncio 2>&1 | tee /tmp/bench_results/batching_heatmap.txt
sudo -n ./python.exe -m profiling.sampling run -r 300khz --jsonl      -o /dev/null              -m test test_asyncio 2>&1 | tee /tmp/bench_results/batching_jsonl.txt

[pablogsal]

This looks fantastic! Thanks a lot @maurycy 🚀 Excellent job

[miss-islington-app]

Thanks @maurycy for the PR, and @pablogsal for merging it 🌮🎉.. I'm working now to backport this PR to: 3.15.
🐍🍒⛏🤖 I'm not a witch! I'm not a witch!

[bedevere-app]

GH-149747 is a backport of this pull request to the 3.15 branch.