feat(dedup): sample column and marginal duplication-ladder columns (#802)

[nh13]

Closes #802. Supersedes #803 (auto-closed when its base branch 786/nhomer/feat-dedup-complexity-metrics was deleted by the #799 squash-merge; GitHub cannot reopen a PR whose base branch was deleted). Same head branch and commit, now rebased onto main.

Toward Picard DuplicationMetrics / dupblaster --stats parity, two dupblaster-derived additions to fgumi dedup's QC output:

sample column

--metrics gains a leading sample column, resolved from the header's unique @RG SM: values (comma-joined), with an optional --sample override. Constant across every row of a run; leads the schema, matching dupblaster/Picard.

Marginal duplication-ladder columns

--duplication-ladder gains per-window columns (window_templates, window_duplicate_fraction) alongside the cumulative ones. The marginal rate isolates each depth band instead of averaging over all prior templates — "usually the more legible view" of the saturation curve (dupblaster's complexity.rs). The windows partition the cumulative total exactly.

Not included

The pair/orphan duplicate breakdown (Picard READ_PAIRS_EXAMINED / READ_PAIR_DUPLICATES / UNPAIRED_*) is a distinct algorithmic change — it requires classifying each template as pair vs orphan during marking — so it's tracked as its own follow-up rather than bundled here.

Verification

New tests: sample resolved from @RG SM, via --sample override, and the empty case (no @RG SM: and no --sample → empty column); ladder window columns pinned exactly (each window_templates equals the increment since the previous snapshot, marginal fractions pinned to num/den, windows summing to the per-library total). The pinned column-order test is updated (sample first). The distinct-sample and override tests now pin exact row identities (["libA","libB","All Reads"] / ["libA","All Reads"]), not just non-empty.

[coderabbitai]

Note

Reviews paused

Use the following commands to manage reviews:

• @coderabbitai resume to resume automatic reviews.
• @coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

• ▶️ Resume reviews
• 🔍 Trigger review

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro

Run ID: 34669677-b56a-4c73-9a70-bd1d1b4fe054

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

_{Comment @coderabbitai help to get the list of available commands.}

[nh13]

@coderabbitai pause

[coderabbitai]

✅ Action performed

Reviews paused.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.47%. Comparing base (e51dc4b) to head (e191081).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #841   +/-   ##
=======================================
  Coverage   94.46%   94.47%           
=======================================
  Files         187      187           
  Lines      115301   115361   +60     
=======================================
+ Hits       108921   108982   +61     
+ Misses       6380     6379    -1     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.