[Python] AST bridge: reject cudaq.dbg.ast aliases and reorderings (#2342)

[mitchdz]

Description:

cudaq.ast is defined as a lazy alias for cudaq.dbg.ast in
_LAZY_SUBMODULES. Because resolveQualifiedName returns obj.__name__
after traversing the attribute chain, cudaq.ast.print_i64(n) was
silently accepted inside kernels — cudaq.ast resolves to cudaq.dbg.ast
at runtime, so obj.__name__ comes back as 'cudaq.dbg.ast' and passes
the devKey guard.

Fix: add isExactCudaqDbgAstCall which walks the literal AST node
structure and requires the chain to be exactly <cudaq_alias>.dbg.ast.<name>.
The devKey check is now conjunctive with this AST check.

Tests cover:

• valid cudaq.dbg.ast.print_i64 and print_f64 calls still work
• cudaq.ast.print_i64 (lazy alias, the specific bug) is now rejected
• dbg.cudaq.ast and ast.cudaq.dbg reorderings are rejected

Fixes #2342

[github-actions]

CUDA Quantum Docs Bot: A preview of the documentation can be found here.

[1tnguyen]

LGTM 👍