As promised, a diagram restating my last post. I've slightly simplified your experimental setup. There are two markers, the "start" marker painted red and the "end" marker painted blue. The rocket is painted purple and passes arbitrarily close to the two markers. In this way I can get rid of your mirror - the red marker is touching the rocket when the light pulse is emitted, so it is received with zero travel time, and the pulse goes direct to the blue marker. This is plotted on the Minkowski diagram below. I've also added the path of the light pulse in yellow.
Basically a Minkowski diagram is a displacement-time graph, with the time axis up the page and with units where the speed of light is one (e.g. seconds for time and light seconds for distance). I've drawn this one in the rest frame of the markers, so their worldlines (blue and red) are vertical (their x coordinate is not changing with coordinate time). Both the rocket (purple worldline) and the light pulse (yellow worldline) are moving to the right, so their worldlines slope right (their x coordinate is increasing with coordinate time). The wordline of the light pulse has a slope of 1 because it is traveling at c - one light second per second. The worldline of the rocket has a slope of 1/0.6 because it is traveling at 0.6c (I only sketched it - don't yell at me if the gradient's wrong).
The rocket passes the red marker at time zero and emits a light pulse which starts the clock at the blue marker when it gets there (when the yellow line crosses the blue line). The rocket stops the clock when it reaches the blue marker (when the purple line crosses the blue line). So what your clock
actually measures is the interval along the blue line between where the yellow line and purple lines cross it. Since the start and stop events are defined by observable physical things, any observer can measure space and time coordinates of each event and calculate the interval between and must come up with the same answer. However, you pre-set your blue clock to d/c. In other words you added the interval between the events "blue line crosses the x axis" and "yellow line meets blue line". But there's nothing special about the first of those two event. There's no way for any observer to spot "blue worldline crosses x axis" because the x-axis is just something I drew on the diagram. And I could have chosen a different simultaneity convention, which would really just be a matter of rotating the x axis. It changes nothing about the worldlines, but renders your choice of d/c meaningless.
The particular choice of simultaneity criterion that I've made (or, Einstein made and I copied) is a very sensible one. But it is a choice, not a physical necessity.