Add truncated normal distribution support

[paulromano]

Description

This PR extends the Normal distribution class to support optional truncation bounds. Normal distributions are sometimes used for source sampling, but their infinite support makes them problematic for biased sampling. OpenMC currently does not let you use a normal distribution as a bias for another distribution with finite support because of the mismatched domains. Truncation allows the normal distribution to be restricted to match the support of the target distribution. A truncated normal is simply specified by passing one or both of the lower and upper arguments. Without these arguments, the untruncated behavior is preserved.

Thanks to @akoen for inspiration!

Checklist

• I have performed a self-review of my own code
• I have run clang-format (version 15) on any C++ source files (if applicable)
• I have followed the style guidelines for Python source files (if applicable)
• I have made corresponding changes to the documentation (if applicable)
• I have added tests that prove my fix is effective or that my feature works (if applicable)

[j-fletcher]

This looks great! It all seems to be working; there's only one quirk I found which is that booleans can be provided as the upper/lower bounds, which are counted as valid alternatives to 0.0 or 1.0 in the support of the distribution, but upon exporting to C++ via XML causes the distribution to silently become un-truncated. I don't think we have to worry too much about it though.

[GuySten]

Looks good to me!