Add relations to simplify convolutions of normally-distributed `RandomVariable`s

[rlouf]

We should add rewrites that we can use to simplify the following expressions for Z_rv, B_rv and C_rv when conditioning:

import aesara.tensor as at

srng = at.random.RandomStream(0)
x_rv = srng.normal(0, 1)
y_rv = srng.normal(0, 2)
Z_rv = x_rv + y_rv # Normal RV

a_rv = srng.normal(0, 1, size=(10,))

B_rv = at.sum(a_rv) # Normal RV
C_rv = at.cumsum(a_rv) # Vector of normal RVs

[larryshamalama]

I can give this a shot. Perhaps I can start with working with Z_rv and see how the logic generalizes to B_rv and C_rv?

[brandonwillard]

I can give this a shot. Perhaps I can start with working with Z_rv and see how the logic generalizes to B_rv and C_rv?

That would be great, thanks!