refactor(stats): shrink and tidy WhichStats

[jqnatividad]

Summary

• Change WhichStats::percentile_list from String to Box<str>. The field is set once from the CLI flag and never mutated, so the unused capacity word is wasted. Shrinks WhichStats from 40 → 32 bytes and is serde-transparent (no change to the on-disk <file>.stats.csv.json cache format; reading old caches still works).
• Move use_weights above percentile_list so all 12 bool fields form a contiguous block with the variable-size Box<str> at the end. Source order has no runtime effect (WhichStats is repr(Rust)); purely a readability nit.

Performance note

This is not a perf win — framed as refactor/style for that reason. Benchmarked on a 514 MB / 1M-row NYC_311_SR_2010-2020-sample-1M.csv (M4 Max, 10 runs each, --force to bypass cache):

Workload	Before	After
stats	1.416 s ± 0.007	1.422 s ± 0.005
stats --everything	2.181 s ± 0.015	2.199 s ± 0.008

Both deltas are inside measurement noise. The theoretical cache-line argument (sub-cache-line spillover from which: WhichStats into Stats' second cache line) doesn't materialize, likely because Stats is #[repr(C, align(64))] and the 8 saved bytes get absorbed into existing tail padding, plus the bool reads are extremely well-predicted per column.

Test plan

• cargo build --locked --bin qsv -F all_features
• cargo clippy -F all_features --bin qsv -- -D warnings
• cargo t stats -F all_features — all 728 stats tests pass
• Verified stats cache JSON file remains structurally compatible (serde matches by name on read)

🤖 Generated with Claude Code

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}