Temperature interpolation bug

[liangjg]

Users from NCSU reported that when using temperature interpolation method in OpenMC, keff shows abnormal behavior: for a pure u235 infinite case, the kinf drops to nearly zero at 610 K, with 600K and 900K library.
I found this bug does not show up in the last stable version "v0.10.0" so it should be introduced in recent updates. Preliminary debugging shows there are unexpected zeros in the "NU_FISSION" array within the lower temperature data (that's why the kinf is closer to the normal value when the target temperature is nearer to the higher temperature). I'll dig into it further. Just let me know if you have any idea about this @paulromano @openmc-dev/committers

[paulromano]

Thanks for taking a preliminary look @liangjg. I'll take a look at the source and see if anything pops out at me. Part of the problem is that our test data currently does not have multiple temperatures, so temperature interpolation is not being tested. We should definitely remedy that so I'll look into setting up a new library for testing.