fix: random number generation

[markus7800]

Fixes #158.

x = [Random.randn(Random.default_rng(), Tracker.TrackedReal{Float64}) for i in 1:10^6]
Distributions.mean(x) 
Distributions.std(x)

now produces a normal sample, and not a uniform sample as before.

[ChrisRackauckas]

Can you add a test for this as well?

[coveralls]

Pull Request Test Coverage Report for Build 6943072974

• 0 of 1 (0.0%) changed or added relevant line in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 72.74%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/lib/real.jl	0	1	0.0%

Totals
Change from base Build 6905129204:	0.0%
Covered Lines:	515
Relevant Lines:	708

---

💛 - Coveralls

[codecov-commenter]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (59b1e5e) 72.74% compared to head (4141bb5) 72.74%.

Files	Patch %	Lines
src/lib/real.jl	0.00%	1 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #159   +/-   ##
=======================================
  Coverage   72.74%   72.74%           
=======================================
  Files           9        9           
  Lines         708      708           
=======================================
  Hits          515      515           
  Misses        193      193           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.