feat(larod): Add safe Rust bindings

[vsem-azamat]

Safe Rust wrapper for the larod C API

Adds a new larod crate with safe Rust bindings around the larod C API. The wrapper covers connection management, device enumeration, model loading/retrieval, tensor allocation, job requests, parameter maps, and error handling.

Review focus

• FFI ownership and cleanup paths for larod-managed pointers.
• Safe API boundaries for tensors, jobs, maps, and file-descriptor access.
• Error handling for null pointers, larod error objects, and fd sentinel values.
• Consistency with lessons from PR feat(vdo): Add safe Rust bindings for VDO API #223: fallible constructors, explicit unsafe contracts, no safe fd lifetime overclaims, and focused tests/docs.

Current status

• Rebased onto current main (ACAP Native SDK 12.11.0, armv7hf support).
• History squashed to one feature commit.
• Since larod is documented as not thread safe, every FFI call goes through one process-wide lock and all handle types are !Send + !Sync (compile-fail doctests). Feedback on this trade-off is welcome: JobRequest::run holds the lock for the duration of inference.
• Tensor fd getters return Result<c_int>; larod has no "unset fd" sentinel, only LAROD_INVALID_FD on error.
• Hardware-gated device-tests have not been run on a device yet.

Local verification

• cargo fmt -p larod -- --check
• git diff --check
• cargo test -p larod
• cargo clippy -p larod --all-targets -- -D warnings
• RUSTDOCFLAGS="-Dwarnings" cargo doc -p larod --document-private-items --no-deps

Issue ticket number and link

• N/A

Checklist before requesting a review

• I have performed a self-review of my own code
• I have verified that the code builds perfectly fine on my local system
• I have added tests that prove my fix is effective or that my feature works
• I have commented my code, particularly in hard-to-understand areas
• I have verified that my code follows the style already available in the repository
• I have made corresponding changes to the documentation