chore: refactor metrics collection

[jonathanpwang]

closes INT-2924

The collect_metrics flag in ExecutionSegment has been renamed to profiling for accuracy because when bench-metrics feature is enabled, at least some minimal set of metrics is always collected.

Turn off flamegraph profiling in benchmarks -- these will be made into a separate on-demand workflow in the near future.

[zlangley]

lgtm

[github-actions]

Benchmarks

group	app_log_blowup	app_total_cells_used	app_total_cycles	app_total_proof_time_ms	leaf_log_blowup	leaf_total_cells_used	leaf_total_cycles	leaf_total_proof_time_ms	max_segment_length	instance	alloc
ecrecover_program	2	15,230,037	(-290,016 [-50.0%]) 290,016	(-10.0 [-0.4%]) 2,354.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc
fibonacci_program	2	51,505,102	(-1,500,137 [-50.0%]) 1,500,137	(+11.0 [+0.2%]) 5,474.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc
regex_program	2	165,028,173	(-4,190,904 [-50.0%]) 4,190,904	(-54.0 [-0.3%]) 15,955.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc
verify_fibair	2	(-61,258 [-0.4%]) 14,529,472	(-201,874 [-50.8%]) 195,420	(-74.0 [-4.8%]) 1,477.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc

Commit: af3dadf

Benchmark Workflow