ci: test against stable python 3.15 and 3.15t

[not-matthias]

TLDR: CI tested Python 3.15 using 3.15.0-beta.1. It now tests the stable 3.15 release, including the free-threaded build.

• Replace the 3.15.0-beta.1 matrix entries with 3.15 and 3.15t. setup-python lists 3.15.0 as stable.
• Select the free-threaded build with the t suffix, like 3.14t, and remove the now-unused architecture matrix field.

[codspeed]

Merging this PR will improve performance by 2.95%

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

ℹ️ 48 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[100]
• test_multiprocessing_map[10]
• test_multiprocessing_map[100000]
• test_multiprocessing_map[10000]
• test_multiprocessing_map[1000]
• test_threadpool_map[1000]
• test_threadpool_map[10000]
• test_threadpool_map[100000]
• test_sleep[0.1]
• test_sleep[0.05]
• …and 38 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

⚡ 25 improved benchmarks
❌ 16 (👁 16) regressed benchmarks
✅ 236 untouched benchmarks

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	WallTime	test_sum_of_squares[sum_of_squares_sum_comprehension_power]	283.9 µs	223 µs	+27.26%
⚡	WallTime	test_sum_of_squares[sum_of_squares_for_loop_power]	220.4 µs	190.5 µs	+15.7%
⚡	WallTime	test_multiprocessing_map[1000]	56.4 ms	49.2 ms	+14.64%
⚡	WallTime	test_tcp_connection[1.1.1.1-53]	978.3 µs	878.8 µs	+11.32%
⚡	WallTime	test_generate_sum_of_subsets_soln[nums0-9]	27.7 µs	26 µs	+6.49%
⚡	WallTime	test_tcp_connection[wikipedia.org-443]	14.7 ms	13.9 ms	+5.11%
⚡	WallTime	test_iir_filter_process	3.1 µs	3 µs	+4.68%
⚡	WallTime	test_hostname_resolution[127.0.0.1]	8.1 µs	7.8 µs	+4.21%
⚡	WallTime	test_make_lowpass	5.6 µs	5.3 µs	+4.02%
⚡	WallTime	test_solve_power_sum[13-2]	7.1 µs	6.8 µs	+3.96%
⚡	WallTime	test_make_highpass	5.6 µs	5.4 µs	+3.72%
⚡	WallTime	test_make_peak	6.5 µs	6.3 µs	+3.53%
⚡	WallTime	test_pipe_communication[1000]	6.2 ms	6 ms	+3.52%
⚡	WallTime	test_minimax[scores0]	7.7 µs	7.5 µs	+3%
⚡	WallTime	test_make_lowshelf	7.5 µs	7.3 µs	+2.67%
⚡	WallTime	test_open_close_fd[100]	916.8 µs	894.4 µs	+2.51%
⚡ ℹ️	Memory	test_multiprocessing_map[100]	3 MB	3 MB	+2.17%
⚡	WallTime	test_sudoku[initial_grid0]	7.8 µs	7.7 µs	+2%
👁	WallTime	test_match_word_pattern[aba-GraphTreesGraph]	103.1 µs	105.4 µs	-2.21%
👁	WallTime	test_multiprocessing_map[10000]	60.4 ms	58.6 ms	+3.22%
...	...	...	...	...	...

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-3838-support-cpython-315 (d0879b0) with master (0d4041b)}

[greptile-apps]

Confidence Score: 5/5

[Low impact] This PR appears safe to merge; no actionable issues were found.

Summary

Updates CI to select stable Python 3.15 builds.

• CI tests stable Python 3.15 in both build modes.

_{Reviews (1) · Last reviewed commit: "ci: test against stable python 3.15 and ..." · Reviewed by Greptile}

[lvaroqui]

LGTM

[GuillaumeLagrange]

lgtm to me