feat: add memory profiling support

[not-matthias]

No description provided.

[avalanche-staging]

Congrats! CodSpeed is installed 🎉

🆕 330 new benchmarks were detected.

You will start to see performance impacts in the reports once the benchmarks are run from your default branch.

Detected benchmarks

• test_multiprocessing_map[1000] (WallTime): 61.1 ms
• test_noop_ellipsis (WallTime): 482 ns
• test_color[graph0-3] (WallTime): 27.8 µs
• test_count_even[count_even_slow] (WallTime): 1.3 ms
• test_sum_of_squares[sum_of_squares_sum_labmda_power] (WallTime): 287 µs
• test_combination_sum[candidates0-8] (WallTime): 11.9 µs
• test_threadpool_map[10000] (WallTime): 368.4 ms
• test_generate_all_combinations[0-0] (WallTime): 1.2 µs
• test_recursive_fibo_20 (WallTime): 5.5 ms
• test_n_queens_solve[board0] (WallTime): 10.6 ms
• test_fs_write[1000] (WallTime): 17.5 µs
• test_array_alloc[1000] (WallTime): 3 µs
• test_iir_filter_set_coefficients[a_coeffs0-b_coeffs0] (WallTime): 917.4 ns
• test_make_allpass (WallTime): 5.6 µs
• test_noop_lambda_decorated (WallTime): 1.4 µs
• test_open_knight_tour[1] (WallTime): 4.1 µs
• test_sleep[0.01] (WallTime): 10.1 ms
• test_hostname_resolution[8.8.8.8] (WallTime): 7.5 µs
• test_combination_lists[4-2] (WallTime): 4 µs
• test_open_close_fd[100] (WallTime): 875.2 µs
• ...

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

---

[GuillaumeLagrange]

do not forget to drop this commit

[GuillaumeLagrange]

same here

[GuillaumeLagrange]

Sorry I meant to request changes

Especially we need to validate the extra stack frame thing in instrumentation by comparing the output to a real prod run!

[codspeed-hq]

Merging this PR will degrade performance by 16.62%

⚡ 11 improved benchmarks
❌ 24 regressed benchmarks
✅ 132 untouched benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	test_multiprocessing_map[10000]	101.3 ms	121.4 ms	-16.62%
⚡	WallTime	test_threadpool_map[100000]	3.9 s	3.8 s	+2.76%
⚡	WallTime	test_fs_read[1000]	857.1 ns	778.9 ns	+10.05%
❌	WallTime	test_generate_all_combinations[5-4]	10.8 µs	11 µs	-2.14%
⚡	WallTime	test_iir_filter_process	3.2 µs	3.1 µs	+3.52%
❌	WallTime	test_hamilton_cycle[graph0]	24.5 µs	25 µs	-2.2%
❌	WallTime	test_process_creation[cat /dev/null]	4.9 ms	5.3 ms	-7.27%
❌	WallTime	test_pipe_communication[10000]	15.3 ms	15.6 ms	-2.28%
❌	WallTime	test_threadpool_map[10]	2.5 ms	2.6 ms	-2.14%
⚡	WallTime	test_recursive_fibo_20	5.5 ms	5.3 ms	+3.62%
❌	WallTime	test_multiprocessing_map[1000]	57 ms	65.5 ms	-12.96%
❌	WallTime	test_threadpool_map[10000]	367.4 ms	377.9 ms	-2.78%
❌	WallTime	test_sum_of_squares[sum_of_squares_sum_labmda_power]	287.2 µs	295.1 µs	-2.68%
❌	WallTime	test_noop_lambda_decorated	1.3 µs	1.3 µs	-4.57%
❌	WallTime	test_color[graph0-3]	27.4 µs	28 µs	-2.09%
❌	WallTime	test_mmap_operation[409600]	130.2 µs	134.1 µs	-2.88%
⚡	WallTime	test_solve_crossword[puzzle0-words0]	2.2 µs	2.1 µs	+2.52%
❌	WallTime	test_tcp_connection[1.1.1.1-53]	834.6 µs	882.6 µs	-5.44%
❌	WallTime	test_process_creation[ls -l]	6.2 ms	6.7 ms	-6.51%
❌	WallTime	test_combination_sum[candidates0-8]	11.8 µs	12.1 µs	-2.65%
...	...	...	...	...	...

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

---

_{Comparing cod-2105-support-memory-profiling-for-python (fc8f559) with master (adee8a1)}