feat(scratch): Add support for installing scratch repositories with uv.lock

[Relm-Arrowny]

fix #1678
Just notice you can do target version: so may be this is not needed.

Changes

• Added use_uv_lock in ScratchRepository
• Modified scratch_install to make use of use_uv_lock to allow uv sync
• Updated Helm schema and tests to reflect these changes

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.34%. Comparing base (a4dbc35) to head (971d74d).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1679   +/-   ##
=======================================
  Coverage   96.34%   96.34%           
=======================================
  Files          46       46           
  Lines        3665     3670    +5     
=======================================
+ Hits         3531     3536    +5     
  Misses        134      134           

☔ 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.

[ZohebShaikh]

use_uv_lock as designed doesn't really deliver on its own premise. Because blueapi installs scratch repos into one shared environment, no single repo's uv.lock tells you what's actually installed in blueapi — it only reflects that repo's own resolution. If repo A opts into use_uv_lock and repo B doesn't (or B has its own lockfile with different pinned versions), B's install can silently override dependencies A's lock was relying on. Per-repo opt-in doesn't prevent this, and worse, it doesn't even reliably work per-repo: if two use_uv_lock repos pin different versions of the same transitive dependency, the outcome depends on install order — it'll work by accident in some configs and break in others.

Proposed fix: move use_uv_lock up to ScratchConfig (not per-repo), so it's an explicit, all-or-nothing decision about the whole scratch environment's install strategy, plus a configurable path to the lockfile to use (defaulting to uv.lock when there's only one scratch repo). For multiple repos, there's no sound way to combine independently-generated lockfiles at install time — a consolidated lockfile covering all scratch repos together needs to be generated in CI at release time instead.