[critical for v3-final release] incorrect spectral profiles of computed polarization quantities

[kswang1029]

Describe the bug
This issue exists in the dev and v3b3 release. I missed this.... 😔

When the polarization dropdown in the spectral profiler is "current", the displayed profile can be updated according to the selection in the animator via the polarization slider. However, if we use the polarization dropdown in the spectral profiler to explicitly select a computed component such as Pangle, the profile is incorrect and seems to be the same as the "current" which is Stokes I. I compared the protobuf message and it seems the backend returned the same profile as the "current". The requirement appears correctly set by the frontend. @markccchiang @TienHao Please help on checking this issue and identify which side the issue sits.

To Reproduce
Steps to reproduce the behavior:

1. load a stokes cube
2. launch a spectral profiler
3. use the polarization dropdown of the spectral profiler to select Pangle
4. the profile stays the same as the "current" was
5. set the polarization dropdown to current and use the animator polarization slide to switch to Pangle to see the correct profile

Expected behavior
Setting the polarization dropdown of the spectral profiler to, for example, Pangle, should result in identical profile as when the polarization dropdown of the spectral profiler is current and the polarization slider of the animator is Pangle.

Screenshots or videos

Screen.Recording.2022-06-09.at.11.23.04.mov

Platform info (please complete the following information):

• OS [e.g. macOS Monterey]: macOS Monterey
• Browser [e.g. chrome, safari, electron app]: chrome and electron app of v3b3
• Browser version [e.g. 22]: 102
• Backend branch [e.g. dev, v3b2 release]: dev and v3b3 release
• Frontend branch [e.g. dev, v3b2 release]: dev and v3b3 release