Closes #3008: Add generator sym entry and stateful uniform distribution

[stress-tess]

This PR (closes #3008) adds a generator sym entry to the symbol table. It also add a stateful uniform distribution. So repeated method calls won't return the same array with a set seed, but a new generator with the same seed will produce the same sequence of random arrays. It updates the methods which draw a from a uniform distribution to use the stateful variant

Example:

>>> rng = ak.random.default_rng(74)

>>> rng.integers(0, 2**32, size=3, dtype="uint")
array([103369013 434448015 1372864948])

>>> rng.uniform(-1.0, 1.0, size=3)
array([-0.31903782753174481 0.35896419541356117 0.45980632831507906])

>>> rng.integers(0, 1, size=5, dtype="bool")
array([False True False True True])

>>> rng.integers(-(2**32), 2**32, size=3, dtype="int")
array([-83456679 -540246299 -3756329859])

# repeat
>>> rng = ak.random.default_rng(74)

>>> rng.integers(0, 2**32, size=3, dtype="uint")
array([103369013 434448015 1372864948])

>>> rng.uniform(-1.0, 1.0, size=3)
array([-0.31903782753174481 0.35896419541356117 0.45980632831507906])

>>> rng.integers(0, 1, size=5, dtype="bool")
array([False True False True True])

>>> rng.integers(-(2**32), 2**32, size=3, dtype="int")
array([-83456679 -540246299 -3756329859])

Its adds tests for this functionality. Note the tests for the legacy functions are copied over from the existing ones in pdarraycreation

[stress-tess]

I forgot there's been changes to the chpl random module in recent versions, I'll need to make this compatible with chpl 1.31 and 1.32

[stress-tess]

I did my best updating the random compat modules, but that was mostly just me loosening the memory management requirements to make the errors go away lol. @jeremiah-corrado seems to have touched those wouldn't mind double checking my work I'd appreciate it

[jeremiah-corrado]

Looks good to me! Just the one question about an extra array copy in the compat modules:

[ajpotts]

A few questions, as marked, but otherwise looks good!

[ajpotts]

Looks good!

[jaketrookman]

Looks good