autotest: print beacon position delta as float, not truncated %u

[peterbarker]

Summary

Adjusts output to add more precision

Classification & Testing (check all that apply and add your own)

• Checked by a human programmer
• Non-functional change
• No-binary change
• Infrastructure change (e.g. unit tests, helper scripts)
• Automated test(s) verify changes (e.g. unit test, autotest)
• Tested manually, description below (e.g. SITL)
• Tested on hardware
• Logs attached
• Logs available on request

Description

BeaconPosition() (also exercised by the Replay Beacon bit) waits for the EKF position to settle within max_delta=1 m of the pre-switch GPS fix and logs the distance with "delta=%u". pos_delta is a float in metres, so %u truncated it: a genuine 1.55 m miss printed as "delta=1", which reads as though it satisfied "want <= 1" yet the test still timed out -- actively misleading when diagnosing the beacon-convergence flake. Print %.2f.

Discovered as part of chasing this outlier down:

[hunt0r]

IMO this comment is unnecessary.
But also OK as-is.

[peterbarker]

Yeah, it really wasn't. Fixed.

[tpwrules]

Does this just make the issue happen at 1.004? Can you think of any reason to limit the number of digits at all? Do you think that is a likely case?

[peterbarker]

Does this just make the issue happen at 1.004? Can you think of any reason to limit the number of digits at all? Do you think that is a likely case?

Pretty sure we have reasonable epsilons here, so we don't need all the digits.

[tpwrules]

I do not see any epsilon in the next line.

I'm not hoping to be difficult, just pointing out that according to my understanding this PR does not fully solve the problem laid out in the description. If you think it's close enough I'm fine, just want to make sure we share an understanding.