Bugfixes may2024

[will-moore]

Fixes for a bug introduced with the un-released vite.js upgrade at #477:

• Channel sliders: for small ranges (e.g. float data where brightest pixel is less than 100) we use a smaller step for the slider and we format the value to float when displaying in the start/end text boxes. We also have logic to prevent the 'end' slider value going less than the 'start' slider. In this bug, I wasn't using the full pixel range to check for using small step and float formatting - therefore the test image (see fake image below) with pixel range of -32768.00 - 72 was using the float formatting etc. Also, the formatting to string step was happening before we compared start < end values, so we were comparing string with number which gave a wrong result (so that when sliding the 'end' value, it would get reset to the same as the start value.

testProjection&sizeX=1952&sizeY=1952&sizeZ=73&sizeC=4&pixelType=int16.fake which is included in https://merge-ci.openmicroscopy.org/web/figure/file/3901/

[pwalczysko]

Tested on merge-ci

• without this fix, the slider on the max side is vanishing when slided to the left, and the max value is wrongly formatted
• with this fix, the slider is working fine and the max value is as expected.

lgtm