The idea is:
The only reason to have a manual trigger is to run an automation from the UI. So when the manual trigger is present and active, the automation should -ALWAYS- start from the manual trigger. Any other behavior is counterintuitive.
If someone wants to test whether their webhook-initiated flow is working right, they should not have a manual trigger present. Then their flow will either used the pinned data in the webhook trigger or wait for a test request.
My use case:
In this flow, there’s a postgres query attached to the manual trigger to reset the flags on the records that will be processed. But as you can see in this screenshot, when I click “test workflow” it starts from the schedule trigger instead of the manual trigger. That is definitely not what I would expect.
I think it would be beneficial to add this because:
Debugging is hard enough without having to wrestle with the debugger!
Any resources to support this?
This behavior isn’t documented.
Are you willing to work on this?
Nobody wants that. 
P.S. Happy Halloween - I like the cobwebs!
