Apply localize minimum coverage before aggregation

[SuhasSrinivasan]

Fixes #651.

Summary

• pass --min-coverage into localize record processing
• retain each bedMethyl record only when valid_coverage >= min_coverage, before offset aggregation
• add an exact end-to-end CLI regression covering below, equal-to, and above-threshold records

Root cause and impact

EntryLocalize::run assigned and logged self.min_coverage, but did not pass it into GenomeRegion::into_localized_mod_counts. The aggregation therefore folded every fetched bedMethyl row into n_valid and n_mod, regardless of the requested threshold.

Filtering must happen per bedMethyl record before aggregation. Applying the threshold after summing offsets would allow multiple individually under-covered loci to combine and pass.

Regression

The test uses two records from the existing indexed bedMethyl resource that localize to the same output bin:

• valid coverage 1, modified count 1
• valid coverage 23, modified count 2

At --min-coverage 1, the exact aggregate remains 24 valid / 3 modified / 12.5%. At --min-coverage 3, the first record is excluded and the exact result is 23 valid / 2 modified / 8.695652%.

Validation

• confirmed the finalized regression fails on the parent revision: threshold 3 still reports 24 / 3 / 12.5 instead of 23 / 2 / 8.695652
• cargo test -p modkit --test test_localize test_localize_min_coverage_filters_before_aggregation -- --exact
• cargo test -p modkit --test test_localize (2 passed)
• cargo test --workspace --quiet -- --test-threads=1
• rustfmt --check --edition 2021 modkit-core/src/localise/subcommand.rs modkit-core/src/localise/util.rs modkit/tests/test_localize.rs
• git diff --check
• installed-versus-patched CLI comparison on the issue fixture confirms thresholds 1 and 3 are identical before the fix and differ exactly as expected after it

The serial workspace mode avoids an existing unrelated collision between pileup integration tests that share a temporary output filename.