Issue in vector widening floating-point arithmetic instructions: subnormal and special cases are not handled

[Shafi10x]

In vector widening floating-point arithmetic instructions, the number is first converted to a double width format. It seems that during the conversion to double-width, implementation does not handle subnormal numbers and special cases. This could potentially lead to unexpected results or errors in certain computations that involve these special cases. Therefore, it's important to ensure that the conversion process includes proper handling of subnormal numbers and special cases.
Example:

If a number is zero in single-precision floating-point format, it should also be zero when converted to double-precision floating-point format but implementation changes the number zero.

[mp-17]

Thanks @Shafi10x,

I will have a look at this! Very much appreciated!

[Shafi10x]

@mp-17 Welcome, I have resolved the issue and have also created a PR for it, which can be viewed at #222. I kindly request you to take a look at it as well. Thank you!