Interpolation debug left turned on

[robertlipe]

./gpsbabel -i gpx -f bigw.gpx -x kalman -o gpx -F bigw-filered.gpx
kalman: [GEN] interpolated point at "Mon Sep 22 14:39:57 2025 GMT" "interpolated2-1" lat: XX.XXXX lon: -YY.YYYY

This is disturbing in two ways.

A) That it's turned on and I didn't notice it.
B) That I didn't notice it. This must mean the test suite never hits it, which is weird because there were definitely some spots that should be interpolated. (The sharp bend exiting the parking lot, turning south, just as the ride ends.)

[tsteven4]

I don't think this else clause can ever be executed (besides being empty).

gpsbabel/kalman.cc

Lines 370 to 373 in 2356727

	} else {
	// This is a marked point but NOT a zinger.
	// We simply skip adding it to new_route_head, but keep last_kept_for_interp
	// so that the next non-deleted point can be interpolated against the previous one.

The existence of the comment leads me to believe you expected a wpt could be marked for deletion but not be a zinger.

is_zinger_deletion and wpt_flags.marked_for_deletion are always set to true in adjacent lines.