Suggest gsl::narrow/gsl::narrow_cast for float/double to integral conversions in ES.46

One of the things ES.46 touches upon is the following:

Enforcement
A good analyzer can detect all narrowing conversions. However, flagging all narrowing conversions will lead to a lot of false positives. Suggestions:

Flag all floating-point to integer conversions (maybe only float->char and double->int. Here be dragons! we need data).

We have recently ran into an issue where directly casting a negative double to an unsigned integral type results in inconsistent code generation when targeting a specific architecture. Therefore, we want to implement a static check that flags all conversions from negative double/float to unsigned integral types and suggests to use gsl::narrow or gsl::narrow_cast. My hope is we can then update the existing implementation of gsl::narrow/gsl::narrow_cast to guard against such conversions by first going from float/double to signed integral type and then convert the result to unsigned, guarding against data loss during the process.

Can we improve the documentation of ES.46 such that it suggests people to use gsl::narrow and gsl::narrow_cast when doing float/double to integral conversions?

Start in CppCoreGuidelines.md at the ES.46 (Res-narrowing) section and review its enforcement guidance on floating-point to integer conversions. Update the documentation to suggest gsl::narrow and gsl::narrow_cast for these conversions, then verify that the ES.46 wording clearly reflects the requested recommendation.