Log commands manually before executing them

[henrymercer]

This PR fixes a bad interaction between #1795 and #1811 which we didn't pick up since the problem only appears when looking at the combination of these two PRs.

In #1811, to make sure we only print the analysis summary once, we invoked the tool runner with silent: true to capture CodeQL's stdout without logging it. We then had our stdout listener forward CodeQL's stdout to the Action's stdout. However we missed another implication of setting silent: true: this disables the log line which prints the command being invoked. This is because we only started running most of our CodeQL invocations through the tool runner in #1795.

With the combination of these two PRs, we stop logging most CodeQL invocations, which reduces the utility of the logs.

This PR starts printing the commands manually to make sure we log them in all cases, including ones where we want to capture the stdout.

Merge / deployment checklist

• Confirm this change is backwards compatible with existing workflows.
• Confirm the readme has been updated if necessary.
• Confirm the changelog has been updated if necessary.

[angelapwen]

Agh.. nice catch.