Fix leak of model_partitions catalog rows when deleting a partitioned model

[nishantmonu51]

• Deleting a partitioned model that had produced output dropped the output table and returned immediately, skipping clearPartitions. The model's rows in the model_partitions catalog table were orphaned forever, since each model gets a fresh PartitionsModelId and clearPartitions is the only caller of DeleteModelPartitions.
• The deletion branch in ModelReconciler.Reconcile now falls through after a successful output drop, so clearPartitions runs on every delete path.
• Added a regression test (TestPartitionsClearedOnDelete) that fails on the old code (all partition rows orphaned) and passes with the fix.

Steps to reproduce:

1. In Rill Developer, create a partitioned incremental model and let it build:

   type: model
   incremental: true
   materialize: true
   partitions:
     sql: SELECT range AS num FROM range(0, 50)
   sql: SELECT {{ .partition.num }} AS num

2. Verify partition rows exist in the catalog: sqlite3 <project>/tmp/meta.db "SELECT COUNT(*) FROM model_partitions;" returns 50.
3. Delete the model file and wait for reconcile (the output table is dropped).
4. Re-run the query from step 2. Before this fix, all 50 rows remain orphaned, and repeated create/delete churn grows the catalog DB unbounded. With the fix, the count returns to 0.

Checklist:

• Covered by tests
• Ran it and it works as intended
• Reviewed the diff before requesting a review
• Checked for unhandled edge cases
• Linked the issues it closes
• Checked if the docs need to be updated. If so, create a separate Linear DOCS issue
• Intend to cherry-pick into the release branch
• I'm proud of this work!

[begelundmuller]

LGTM