perf(translate): fix slow translation startup — CJK estimation, output budget, provider token config

Root cause: batch sizing underestimated CJK token density (1.5→1.0 chars/token)
and ignored output budget as primary constraint, causing cascading finish_reason=length.

Changes:
- _token_budget.py: CJK_RATIO 1.5→1.0, OTHER_RATIO 2.2→1.8, safety factors 0.75/0.70
- _token_budget.py: new _compute_max_rows_by_output() — output budget is PRIMARY constraint
- _batch_sizer.py: resolve_provider_config() with DB-level context_window/max_output_tokens
- _batch_sizer.py: INPUT_SAFETY_FACTOR applied, max_rows_by_output used as row cap
- _llm_http.py: log actual usage.prompt_tokens/.completion_tokens from provider
- _llm_call.py: retry only missing rows after finish_reason=length (save partial result)
- models/llm.py + schema: provider-level context_window / max_output_tokens (nullable)
- services/llm_provider.py: get_provider_token_config() helper
- Alembic migration: add columns to llm_providers
- Svelte ProviderConfig: collapsible Advanced: Token Limits section
- 12 new tests (token budget, batch sizer, provider config)
- All 492 tests pass

backend/alembic/versions/20260603_add_token_limits_to_llm_providers.py

@@ -0,0 +1,48 @@
+"""Add context_window and max_output_tokens to llm_providers
+
+Revision ID: a1b2c3d4e5f6
+Revises: f1a2b3c4d5e6
+Create Date: 2026-06-03
+
+Add token window configuration to LLM provider records:
+- context_window: total context window in tokens (nullable)
+- max_output_tokens: max output tokens limit (nullable)
+Both NULL = use PROVIDER_DEFAULTS fallback from model name.
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = "a1b2c3d4e5f6"
+down_revision: Union[str, None] = "f1a2b3c4d5e6"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "llm_providers",
+        sa.Column(
+            "context_window",
+            sa.Integer(),
+            nullable=True,
+            comment="Total context window in tokens. NULL = auto-detect from model name",
+        ),
+    )
+    op.add_column(
+        "llm_providers",
+        sa.Column(
+            "max_output_tokens",
+            sa.Integer(),
+            nullable=True,
+            comment="Max output tokens limit. NULL = auto-detect from model name",
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("llm_providers", "max_output_tokens")
+    op.drop_column("llm_providers", "context_window")