Fix initialization.md documentation: steady-state and observable initialization example

[ChrisRackauckas-Claude]

Description

The "More Features of the Initialization System: Steady-State and Observable Initialization" section in docs/src/tutorials/initialization.md has disabled examples due to compatibility issues.

Problem

The examples demonstrating steady-state initialization (setting derivatives to zero) and observable initialization are currently disabled due to a compatibility issue with the initialization system and the current ModelingToolkit stack.

Location

docs/src/tutorials/initialization.md - lines 496-542

Expected behavior

The following patterns should work:

1. Steady-state initialization - Starting an ODE at steady state by setting derivatives to zero:

prob = ODEProblem(simpsys, [D(x) => 0.0, D(y) => 0.0], tspan, guesses = [x => 1, y => 1])

2. Observable initialization - Using an observable's initial condition directly:

prob = ODEProblem(simpsys, [D(x) => 0.0, z => 2.0], tspan, guesses = [x => 1, y => 1])

Related code (Lotka-Volterra example)

@variables x(t) y(t) z(t)
@parameters α=1.5 β=1.0 γ=3.0 δ=1.0

eqs = [D(x) ~ α * x - β * x * y
       D(y) ~ -γ * y + δ * x * y
       z ~ x + y]

@named sys = System(eqs, t)
simpsys = mtkcompile(sys)