Add typed PyTuple sequence wrapper

[cjdoris]

Motivation

• Provide a typed Julia wrapper for Python tuples and sequence-like objects so tuple element types/lengths declared in Julia (T<:Tuple) are respected and conversions behave consistently.
• Support both statically-known tuple types and vararg/unconstrained tuple types, using the type when it encodes length/element types and falling back to the wrapped Python object's length otherwise.

Description

• Add PyTuple{T<:Tuple} and constructor in src/API/types.jl and export it from src/API/exports.jl.
• Implement src/Wrap/PyTuple.jl with ispy, Py, pyconvert_rule_pytuple, IteratorSize/IteratorEltype, eltype, length logic (use the type when fixed else query pylen), size, bounds-checked getindex that converts elements using fieldtype(T,i), iterate, and Base.Tuple materialization via pyconvert.
• Register a normal-priority pyconvert rule for Python sequences in src/Wrap/Wrap.jl via pyconvert_add_rule("collections.abc:Sequence", PyTuple, pyconvert_rule_pytuple, priority).
• Add tests covering construction, conversion, indexing, iteration, tuple materialization and a variety of tuple shapes in test/Wrap.jl and test/Convert.jl.

Testing

• Ran the full Julia test suite with julia --project -e 'using Pkg; Pkg.test()' and the suite passed in this environment (all package tests exercised; PythonCall tests passed).
• Ran coverage with julia --project -e 'using Pkg; Pkg.test(coverage=true)' and the new src/Wrap/PyTuple.jl code was exercised by tests.
• Ran the Python tests with uv run pytest -s --nbval ./pytest and all Python tests passed.

---

Codex Task