Augment raymath.h with useful functions

[saccharineboi]

I use these functions heavily during gamedev so I thought they could be useful for others too:

• FloatEquals: Check whether two given floats are almost equal
• Vector{n}Invert: Invert the given vector
• Vector{n}Clamp: Clamp the components of the vector between min and max values specified by the given vectors
• Vector{n}ClampMagnitude: Clamp the magnitude of the vector between two given min and max values
• Vector{n}Equals: Check whether two given vectors are almost equal

Additionally Vector3{Up,Down,Left,Right,Forward,Backward} are added for convenience.

Vector3Refract refracts a vector given a normal and a refractive index.

QuaternionEquals checks whether two given quaternions are almost equal.

Furthermore, a macro named "EPSILON" with a value 1e-6 is added (used in floating-point comparison functions), and stdbool.h is included for functions that return boolean values.

[CrezyDud]

rename Vector2ClampMagnitude and Vector3ClampMagnitude to Vector2ClampValue and Vector3ClampValue
for consistency

[raysan5]

@saccharineboi I think some of those functions are unneded and just add a level of undirection not required. i.e users cand define required Vector3 up/down/back... directly, no need to provide functions for everything.

[raysan5]

@saccharineboi thanks for the review.

[planetis-m]

I don't know if it's intentional or an oversight but missing from this PR is MatrixEquals.

[raysan5]

@planetis-m Could you list a use case for that function?

Actually, I'm already considering removing all the *Equals() functions, the bool include broke raylib.

[saccharineboi]

I don't know if it's intentional or an oversight but missing from this PR is MatrixEquals.

Intentional. In contrast to Vector{n}Equals which is useful for checking degenerate vectors, MatrixEquals is almost never used. Though it could be added for completeness' sake.

@raysan5 what do you mean it broke raylib? Was it because of stdbool.h? Instead of removing those functions, you could change return types to int

[raysan5]

@saccharineboi I already changed return types to int, I'm considering the real use case for those functions. As it happens with many raylib functions, they looks like a good idea to have but nobody is actually using them ever.

For that reason I try to be very careful with raylib additions. Let's see if anybody will ever use those functions...