ci: migrate to github default arm runners

[lvaroqui]

No description provided.

[greptile-apps]

Confidence Score: 5/5

[High impact] The runner change appears safe to merge.

Summary

Switches the ARM release build from the custom runner to GitHub’s ubuntu-24.04-arm.

• Linux ARM wheels now build on GitHub’s default Ubuntu ARM runner.

_{Reviews (1) · Last reviewed commit: "ci: migrate to github default arm runner..." · Reviewed by Greptile}

[codspeed]

Merging this PR will improve performance by 0.05%

⚠️ 6 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• test_generate_all_permutations[sequence0]
• test_generate_all_permutations[sequence1]
• test_recursive_cached_fibo_100
• test_n_queens_solve[board0]
• test_generate_all_subsequences[sequence1]
• test_generate_all_subsequences[sequence0]

ℹ️ 46 benchmarks have no flame graph

The flamegraph could not be generated because we failed to resolve the symbols in the profiling trace. This happens when we fail to harvest information about language frames.

• test_multiprocessing_map[10]
• test_multiprocessing_map[100]
• test_multiprocessing_map[1000]
• test_multiprocessing_map[100000]
• test_multiprocessing_map[10000]
• test_threadpool_map[1000]
• test_process_creation[echo hello]
• test_process_creation[ls -l]
• test_threadpool_map[10000]
• test_process_creation[cat /dev/null]
• …and 36 more

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 20 improved benchmarks
❌ 17 (👁 17) regressed benchmarks
✅ 240 untouched benchmarks

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡ ℹ️	Memory	test_multiprocessing_map[10]	4.4 MB	3.1 MB	+43.92%
⚡	WallTime	test_tcp_connection[wikipedia.org-443]	19.4 ms	14 ms	+39.05%
⚡ ℹ️	Memory	test_multiprocessing_map[100]	3.4 MB	3 MB	+14.28%
⚡	WallTime	test_multiprocessing_map[100]	56.2 ms	49.5 ms	+13.48%
⚡ ℹ️	Memory	test_multiprocessing_map[1000]	3 MB	2.7 MB	+13.08%
⚡	WallTime	test_noop_pass_decorated	721.3 ns	657.9 ns	+9.64%
⚡	WallTime	test_tcp_connection[1.1.1.1-53]	955.7 µs	881.9 µs	+8.37%
⚡	WallTime	test_sum_of_squares[sum_of_squares_sum_comprehension_power]	239.6 µs	221.8 µs	+8%
⚡	WallTime	test_pipe_communication[100]	6.4 ms	6.1 ms	+5.34%
⚡	WallTime	test_sum_of_squares[sum_of_squares_sum_labmda_product]	293.5 µs	278.9 µs	+5.25%
⚡	WallTime	test_match_word_pattern[aba-GraphTreesGraph]	107.8 µs	103.3 µs	+4.36%
⚡	Simulation	test_iterative_fibo_10	41.3 µs	39.8 µs	+3.93%
⚡	WallTime	test_tcp_connection[8.8.8.8-53]	696.3 µs	672.7 µs	+3.5%
⚡	WallTime	test_open_close_fd[100]	897.2 µs	873.1 µs	+2.76%
⚡	WallTime	test_generate_all_subsequences[sequence0]	190.4 µs	185.4 µs	+2.7%
⚡ ℹ️	Memory	test_multiprocessing_map[100000]	5.9 MB	5.8 MB	+2.67%
⚡	WallTime	test_make_lowshelf	7.6 µs	7.4 µs	+2.42%
⚡	WallTime	test_make_peak	6.7 µs	6.5 µs	+2.4%
⚡	WallTime	test_get_valid_pos[pos0-4]	2.9 µs	2.8 µs	+2.38%
⚡	WallTime	test_pipe_communication[10]	7.9 ms	7.8 ms	+2.06%
...	...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Tip

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing cod-3837-migrate-from-codspeedhq-arm64-ubuntu-204-github-runners-to (4c68bfd) with master (14b0bb9)}

[GuillaumeLagrange]

@lvaroqui for non tricky PRs like this, one reviewer is fine IMO, unless you explicitly need the input from two different persons