Fix documentation build errors

- low_level_2.md, param_estim.md: Replace deprecated `defaults` kwarg
  with `initial_conditions` for PDESystem constructor (MTK 11 change)
- pino_ode.md: Fix get_trainset to return 2D trunk input instead of 3D
  (NeuralOperators 0.6+ requires 2D)
- param_estim.md: Update ComponentArrays.jl link from jonniedie to SciML

Fixes makedocs errors: [:linkcheck, :example_block]

Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: ChrisRackauckas

docs/src/tutorials/low_level_2.md

@@ -116,7 +116,7 @@ discretization = NeuralPDE.BayesianPINN([chain],
     [x, t],
     [u(x, t)],
     [α],
-    defaults = Dict([α => 0.5]))
+    initial_conditions = Dict([α => 0.5]))
 
 sol1 = ahmc_bayesian_pinn_pde(pde_system,
     discretization;

docs/src/tutorials/param_estim.md

@@ -80,7 +80,7 @@ three arguments:
   - the hyperparameters `p` .
 
 For a Lux neural network, the composed function will present itself as having θ as a
-[`ComponentArray`](https://github.com/jonniedie/ComponentArrays.jl)
+[`ComponentArray`](https://github.com/SciML/ComponentArrays.jl)
 subsets `θ.x`, which can also be dereferenced like `θ[:x]`. Thus, the additional
 loss looks like:
 
@@ -98,7 +98,7 @@ discretization = NeuralPDE.PhysicsInformedNN([chain1, chain2, chain3],
     NeuralPDE.QuadratureTraining(; abstol = 1e-6, reltol = 1e-6, batch = 200), param_estim = true,
     additional_loss = additional_loss)
 @named pde_system = PDESystem(eqs, bcs, domains, [t], [x(t), y(t), z(t)], [σ_, ρ, β],
-    defaults = Dict([p .=> 1.0 for p in [σ_, ρ, β]]))
+    initial_conditions = Dict([p => 1.0 for p in [σ_, ρ, β]]))
 prob = NeuralPDE.discretize(pde_system, discretization)
 callback = function (p, l)
     println("Current loss is: $l")

docs/src/tutorials/pino_ode.md

@@ -56,7 +56,7 @@ function get_trainset(bounds, tspan, number_of_parameters, dt)
     p_ = [range(start = b[1], length = number_of_parameters, stop = b[2]) for b in bounds]
     p = vcat([collect(reshape(p_i, 1, size(p_i, 1))) for p_i in p_]...)
     t_ = collect(tspan[1]:dt:tspan[2])
-    t = collect(reshape(t_, 1, size(t_, 1), 1))
+    t = reshape(t_, 1, size(t_, 1))
     (p, t)
 end