Add UniRng a random scalar generator

[moCello]

Resolves: #129

[vlopes11]

As mentioned on the JubJub PR, I think adding a new RNG will increase type complexity & degradate performance (i.e. we need to copy the numbers into an array, and then read the numbers from the same array to generate a scalar). I suggest just adding it as a function of the scalar

[moCello]

After extensive discussions we decided to not change the random scalar generation because it was deemed secure enough for our purposed:

We will use this paper to determine whether the current implementation of the random scalar implementation as done in the Field trait implmentation is cryptographically safe.

In the current implementation of random, we sample a random element with 512 bits, and reduce it by the modulo of the field of BlsScalar to generate a valid scalar.

For BlsScalar we have:

p := 0x73eda753299d7d483339d80809a1d80553bda402fffe5bfeffffffff00000001
p = 52435875175126190479447740508185965837690552500527637822603658699938581184513

and

s := 2^512
s = 13407807929942597099574024998205846127479365820592393377723561443721764030073546976801874298166903427690031858186486050853753882811946569946433649006084096

To check whether the above approach is secure, we do the following:

For p < s calculate r and m so that

s = rp + m

This leads to:

r := 255699135089535202043525422716183576215815630510683217819334674386498370757523

and

m := s - p * r
m = 3294906474794265442129797520630710739278575682199800681788903916070560242797

Now we can check whether:

r > 2^64 * root(m/p)
r > 2^64 * root(0.06283687387289490447)
r > 18446744073709551616 * 0.25067284231223554

This clearly holds true, which means that we fall in the first conversion of the paper above, and can safely sample an element from the field over 2^512 (which is an array of 64 bytes) and take it modulo p.

This concludes that we don't need to make any adjustments to the current implementation of random scalar generation.