refactor: centralize nanoseconds constant

[nightcityblade]

Summary

Define NANOSECONDS_PER_SECOND once in hflow.format and import it from resample and testing.

Closes #63.

Why

The identical time-unit constant was maintained in two modules. Moving it to the established shared format-constants module removes that duplication without changing behavior.

Validation

• uv sync --locked --all-extras — passed.
• uv run ruff check --fix — passed.
• uv run ruff format — 88 files left unchanged.
• uv run ty check — passed.
• uv run pytest -q in Debian with Python 3.11 and FFmpeg — 307 passed, 3 skipped.
• rg -n "NANOSECONDS_PER_SECOND = " src/hflow/ — exactly one definition.

Checklist

• N/A — no business logic changed; the full existing test suite covers the moved imports.
• N/A — no behavior, flags, formats, or requirements changed.
• I ran uv run ruff check --fix, uv run ruff format, and uv run ty check.
• I ran the relevant pytest suite.
• I did not add recordings, generated media, credentials, private URLs, or runtime artifacts.
• I preserved stored-data compatibility or documented an explicit version change.

[kstonekuan]

Thank you @nightcityblade, and welcome! Exactly the change the issue asked for, with the verification command in the description so the result is easy to confirm.

What I validated locally:

• Full quality gate is clean (ruff check, ruff format --check, ty check) and all 307 tests pass.
• Confirmed rg -n "NANOSECONDS_PER_SECOND = " src/hflow/ now reports exactly one definition.
• Checked that importing hflow.format from resample.py introduces no import cycle. format.py is leaf-level by design, which is what makes it the right home.

One fixup pushed to your branch (f0521b1): the constant landed between the grid-resampling policy comment and the RESAMPLE_POLICY_VERSION that comment documents ("bump this when the grid or selection semantics change"), so I moved it above that block with its own note. That placement was my issue text's fault for saying "near RESAMPLE_POLICY_VERSION, which already describes the grid semantics"; your reading of it was reasonable.

Merging now. If you want another, the open good first issues have plenty left, and #62 is the close cousin of this one (same shape, different constants). Issues with an assignee are taken; everything else is fair game. We also hang out on Discord.