Update v2.0 documentation with augment=true and Flow(ocp) features

- Add CTFlows enhancements section in CHANGELOG.md
- Document augment=true feature for automatic costate computation
- Document Flow(ocp) direct creation from control-free problems
- Add comprehensive CTFlows Features section in BREAKING.md
- Include migration examples for augmented Hamiltonian approach
- Fix linting issues (duplicate headings, list formatting)
- Maintain consistent technical tone throughout documentation

Author: ocots

BREAKING.md

@@ -10,7 +10,8 @@ Version 2.0.0 represents a major architectural redesign of OptimalControl.jl, in
 - **GPU/CPU parameter system** for heterogeneous computing
 - **Advanced option routing** with introspection and disambiguation tools
 - **New solver integrations** (Uno, MadNCL)
-- **Control-free problems** support
+- **Control-free problems** support with augmented Hamiltonian approach
+- **CTFlows enhancements** with `augment=true` and direct OCP flow creation
 - **Modernized reexport system** using `@reexport import`
 
 ## Removed Functions
@@ -172,7 +173,7 @@ The old functional approach is no longer supported.
 
 These features are new in v2.0.0 but don't break existing code:
 
-### Control-Free Problems
+### Control-Free Problems Support
 
 Support for optimal control problems without control variables:
 
@@ -220,6 +221,45 @@ using CUDA, MadNLPGPU
 sol = solve(ocp, :collocation, :exa, :madnlp, :gpu)
 ```
 
+## CTFlows Features
+
+### Control-Free Problems
+
+v2.0.0 introduces comprehensive support for control-free problems (optimal control without control variables) with enhanced CTFlows integration:
+
+**Augmented Hamiltonian approach:**
+
+```julia
+# v1.1.6: Manual augmented Hamiltonian construction
+function H_aug(t, x_, p_)
+    x, λ = x_
+    p, _ = p_
+    return H(t, x, p, λ)
+end
+f = Flow(Hamiltonian(H_aug))
+
+# v2.0.0: Direct OCP flow creation
+f = Flow(ocp)
+```
+
+**Automatic costate computation:**
+
+```julia
+# v2.0.0: augment=true automatically computes p_λ(tf)
+function shoot!(s, p0, λ)
+    _, px_tf, pλ_tf = f(t0, x0, p0, tf, λ; augment=true)
+    s[1] = px_tf  # p(tf) = 0
+    s[2] = pλ_tf  # p_λ(tf) = 0
+end
+```
+
+**Mathematical framework:**
+
+- Complete augmented system dynamics with proper transversality conditions
+- Automatic handling of Lagrange costs: $p_\lambda(t_f) = 0$
+- Automatic handling of Mayer costs: $p_\omega(t_f) = -2\omega$
+- Initial conditions: $p_\lambda(t_0) = 0$ by construction
+
 ## Dependency Updates
 
 v2.0.0 requires updated versions of CTX packages:

CHANGELOG.md

@@ -55,6 +55,10 @@ Versions follow [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   - Optimal control problems without control variables
   - Optimization of constant parameters in dynamical systems
   - Full integration with solve pipeline
+  - **Augmented Hamiltonian approach**: `augment=true` feature in CTFlows for automatic costate computation
+  - **Simplified flow creation**: `Flow(ocp)` directly creates Hamiltonian flow from control-free problems
+  - **Mathematical framework**: Complete transversality conditions for variable parameters
+  - **Documentation**: Comprehensive examples with exponential growth and harmonic oscillator
 
 - **New solvers**:
   - **Uno**: CPU-only nonlinear optimization solver (methods with `:uno`)
@@ -78,6 +82,12 @@ Versions follow [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   - Organized by source package (ctbase.jl, ctdirect.jl, ctflows.jl, ctmodels.jl, ctparser.jl, ctsolvers.jl)
   - Cleaner separation between imported and exported symbols
 
+- **CTFlows enhancements**:
+  - **Augmented Hamiltonian computation**: `augment=true` automatically computes costates for variable parameters
+  - **Direct OCP flow creation**: `Flow(ocp)` creates Hamiltonian flow without manual Hamiltonian definition
+  - **Transversality conditions**: Automatic handling of $p_\lambda(t_f) = 0$ for Lagrange costs and $p_\omega(t_f) = -2\omega$ for Mayer costs
+  - **Mathematical rigor**: Complete augmented system dynamics with proper initial conditions
+
 - **Strategy registry system**:
   - `StrategyRegistry` with metadata for all strategies
   - `StrategyMetadata` with id, options, and parameter support