Luna-Flow/luna-poly 0.2.0

The library has one shared capability layer and two explicit execution models:

• core provides shared types, capability traits, and functional operation records.
• immut provides persistent value semantics.
• mutable provides explicit setters and _inplace operations.

Concrete implementations live in mirrored subpackages:

• immut/dense, immut/term, immut/sparse, immut/context
• mutable/dense, mutable/term, mutable/sparse, mutable/context

The top-level immut and mutable packages are facades that re-export core capabilities plus their concrete implementations. Both facades expose dense univariate, sorted-term multivariate, and ordered-map sparse polynomials. Canonical form and natural-power behavior are shared across both packages.

Use UnivariatePolynomial, MultivariatePolynomial, ContextualPolynomial, and MutablePolynomial when an algorithm only needs a capability boundary. Use Type::ops() records when functional-style generic code needs construction, conversion, or algebraic operations without matching on the storage representation.

The 0.2.0 API also adds Variable, VariableContext, and ContextPolynomial for named-variable workflows. Context polynomials bind a variable table to either term-array or sparse storage and support indexed and named evaluation. See the root README and generated package interfaces for the current API.

Package Documents

• immut/api.md: value-semantics API reference for DensePolynomial, TermPolynomial, SparsePolynomial, and ExponentVector.
• immut/tutorial.md: examples for dense, multivariate, and sparse immutable polynomials.
• immut/design.md: canonical-form and representation invariants.
• mutable/api.md: mutable API reference, including setters and _inplace methods.
• mutable/tutorial.md: examples for in-place updates and conversions.
• mutable/design.md: mutation boundaries and alignment with immut.