chore(ci): Display times in compilation and execution reports only with seconds

[vezenovm]

Description

Problem*

In #6874 we moved to computing an average of multiple runs of nargo compile and nargo execute. The noir-bench-report still expects our time input to have the format of the output from the time command (e.g. [0-9]+m[0-9]+.[0-9]+s). In that PR we just prepended 0m as to not to have to convert back to minutes. This suffix clutters the report output.

Summary*

In this PR I decided to just keep with total seconds rather than converting back to minutes and seconds. It is simpler to work with one unit in my opinion. If we have reports that really need to be represented in minutes, we can address that at the time (if we have reports of that size it most likely means we have larger issues than how we display our reports).

Updates to the commit in this PR noir-lang/noir-bench-report#6

Additional Context

Documentation*

Check one:

• No documentation needed.
• Documentation included in this PR.
• [For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist*

• I have tested the changes locally.
• I have formatted the changes with Prettier and/or cargo fmt on default settings.

[github-actions]

Peak Memory Sample

Program	Peak Memory
keccak256	78.56M
workspace	123.39M
regression_4709	423.67M
ram_blowup_regression	1.58G
private-kernel-tail	207.23M
private-kernel-reset	720.82M
private-kernel-inner	292.44M
parity-root	172.24M

[github-actions]

Execution Sample

Program	Execution Time	%
sha256_regression	0.619s	Infinity%
regression_4709	0.409s	Infinity%
ram_blowup_regression	4.356s	Infinity%
rollup-base-public	20.905s	Infinity%
rollup-base-private	19.296s	Infinity%
private-kernel-tail	0.691s	Infinity%
private-kernel-reset	1.488s	Infinity%
private-kernel-inner	0.988s	Infinity%
parity-root	0.537s	Infinity%

[github-actions]

Compilation Sample

Program	Compilation Time	%
sha256_regression	1.485s	Infinity%
regression_4709	0.840s	Infinity%
ram_blowup_regression	16.183s	Infinity%
rollup-base-public	132.495s	Infinity%
rollup-base-private	108.628s	Infinity%
private-kernel-tail	1.106s	Infinity%
private-kernel-reset	8.344s	Infinity%
private-kernel-inner	2.338s	Infinity%
parity-root	0.775s	Infinity%
noir-contracts	97.779s	Infinity%

[vezenovm]

Although the diff percentage shows Infinity%, the next report will display correct numbers based off the report generated on the parent branch #6881.

[jfecher]

Infinity%

I'm not sure, may have to block this until perf improves. Joking of course 👍