Refactor: extract lib.rs into focused sub-modules, add --output-dir CLI option

- Split monolithic src/lib.rs (~640 lines) into 7 sub-modules:
  types.rs, context.rs, user_data.rs, translate.rs, http.rs,
  csv_util.rs, test_utils.rs
- Add configurable --output-dir CLI option for CSV output files
- Implement atomic file writes (write-to-temp + rename) for all CSVs
- Change compute_car_counts() to return CarCounts directly (not Result)
- Add test-utils feature gate for shared test helpers
- Add HTTP mock tests with mockito (6 tests)
- Add proptest tests for compute_car_counts (3 property tests)
- Update integration tests to use shared test utilities
- Update documentation (API.md, ARCHITECTURE.md, DATA_STRUCTURES.md)

src/commands/best.rs

@@ -9,11 +9,11 @@ use crate::commands::{load_my_cars_or_warn, run_analysis};
 use crate::output::{write_best_csv, write_very_best_csv};
 
 /// Executes the best command: generates best.csv and very_best.csv.
-pub async fn run_best(my_cars_path: &str, cars_db_path: &str, times_dir: &str, min_count: usize) -> Result<(), AppError> {
+pub async fn run_best(my_cars_path: &str, cars_db_path: &str, times_dir: &str, output_dir: &str, min_count: usize) -> Result<(), AppError> {
     let my_data = load_my_cars_or_warn(my_cars_path)?;
     let (_ctx, counts) = run_analysis(&my_data.owned, cars_db_path, times_dir).await?;
-    write_best_csv(&counts)?;
-    write_very_best_csv(&counts, min_count)?;
+    write_best_csv(output_dir, &counts)?;
+    write_very_best_csv(output_dir, &counts, min_count)?;
     println!("Written: best.csv, very_best.csv");
     Ok(())
 }