s_thyme: rounding error when creating the time grid

[spribitzer]

In very rare cases a bug appears when using indirect dimensions. This mostly happens in combination with moving pulses. It is possible, that the total length of detection (in microseconds) stays constant but the number of detected points change. This is illustrated in the following example, where signals are detected during two consecutive events:

1st Acquisition point of indirect dimension:
tau1 = 0.5 μs, tau2 = 0.5 μs, dt = 0.2 μs
total detection time = 0.5 + 0.5 = 1 μs
total number of data points = 2 + 2 = 4

2nd Acquisition point of indirect dimension:
tau1 = 0.4 μs, tau2 = 0.6 μs, dt = 0.2 μs
total detection time = 0.4 + 0.6 = 1 μs
total number of data points = 2 + 3 = 5

Initially, s_thyme notices that the total length of detection is constant, and creates an n-dimensional vector array to store all the traces. However, during evaluation of the second acquisition point, the size of the array won't be able to accommodate the trace, which throws an error.

[spribitzer]

improved handling with commit 62d01c5.
This fix appears to not work with all MATLAB versions though.

[spribitzer]

Bug fixed with commit 68cfc47. Tests added with commit 98aa7ec.