Verbosity of GPT-5.2 since 0.78

What version of Codex is running?

0.77, 0.78, 0,79

What subscription do you have?

Pro

Which model were you using?

gpt-5.2

What platform is your computer?

WSL

What issue are you seeing?

What I liked since the introduction of GPT-5.2 together with Codex CLI v0.77 for example is that GPT-5.2 was not as noisy and verbose while reasoning and working.

Starting with v0.78 and also v0.79, I can see that the verbosity of thinking is back again. Downgrading to 0.77 confirms that something has changed in 0.78.

If the behavior in 0.77 was in fact caused by a bug, it would at least be great if the verbosity of its reasoning process could be controlled somehow. I'm not talking about the summary - its the in-flight output.

I also tried to pinpoint which commit caused the changed behavior between 0.77 and 0.78 but had no success so far.

What steps can reproduce the bug?

Use 0.77, let it work on something, observe how much it outputs during work.
Use 0.78+, start the same task and observe the difference.

What is the expected behavior?

Less verbose reasoning output as in 0.77, or at least a configuration option.

Additional information

_No response_