fix(ci): restore daily-branch pricing entries removed by drift + add router docs

Base drift commit synced model_prices JSON from main, but main lacks 263
model entries that the daily branch's tests rely on (xai/grok-* models,
claude-3-7-sonnet, claude-3-haiku, gemini-2.0-flash, command, dall-e-3,
text-moderation-stable). Merge those entries back from the daily base
to fix ~8 failing test shards. Sync backup JSON to match. Add router
settings reference docs to satisfy documentation validation test.

Refs: #34444
Signed-off-by: Bartok9 <259807879+Bartok9@users.noreply.github.com>

docs/my-website/docs/proxy/config_settings.md

@@ -0,0 +1,117 @@
+# LiteLLM Router Configuration Settings
+
+Reference documentation for all `Router` configuration parameters.
+
+### router_settings - Reference
+
+| param | type | default | description |
+|-------|------|---------|-------------|
+| model_list | list | [] | List of model deployments for routing |
+| assistants_config | dict | None | Config for OpenAI assistants integration |
+| search_tools | dict | None | Search tool configurations |
+| guardrail_list | list | None | Guardrail configurations |
+| redis_url | str | None | Redis connection URL |
+| redis_host | str | None | Redis host address |
+| redis_port | int | None | Redis port number |
+| redis_password | str | None | Redis password |
+| redis_db | int | None | Redis database number |
+| cache_responses | bool | None | Enable response caching |
+| cache_kwargs | dict | None | Cache configuration kwargs |
+| caching_groups | bool | None | Enable caching for router groups |
+| caching | bool | None | Enable caching |
+| client_ttl | int | None | Client time-to-live |
+| polling_interval | float | None | Polling interval for health checks |
+| default_priority | int | None | Default priority for deployments |
+| num_retries | int | None | Number of retries for requests |
+| max_fallbacks | int | None | Maximum number of fallbacks |
+| retry | int | None | Retry count |
+| default_max_retries | int | None | Default max retries |
+| timeout | float | None | Request timeout |
+| retry_policy | str | None | Retry policy name |
+| stream_timeout | float | None | Stream timeout |
+| default_litellm_params | dict | None | Default LiteLLM parameters |
+| default_max_parallel_requests | int | None | Max parallel requests (general) |
+| max_parallel_requests | int | None | Max parallel requests per model |
+| max_parallel_add_sequence | int | None | Max parallel add sequence |
+| set_verbose | bool | None | Enable verbose logging |
+| debug_level | str | None | Debug level |
+| default_fallbacks | list | None | Default fallback deployments |
+| fallbacks | list | None | Fallback configurations |
+| context_window_fallbacks | list | None | Context window fallback configs |
+| content_policy_fallbacks | list | None | Content policy fallback configs |
+| model_group_alias | dict | None | Model group alias mappings |
+| enable_pre_call_checks | bool | None | Enable pre-call checks |
+| enable_tag_filtering | bool | None | Enable tag filtering |
+| tag_filtering_match_any | bool | None | Tag filtering match any |
+| plugins | bool | None | Enable plugins |
+| retry_after | int | None | Retry after status codes |
+| allowed_fails | int | None | Allowed fails count |
+| allowed_fails_policy | str | None | Allowed fails policy |
+| cooldown_time | float | None | Cooldown time |
+| disable_cooldowns | bool | None | Disable cooldowns |
+| routing_strategy | str | None | Routing strategy |
+| optional_pre_call_checks | bool | None | Optional pre-call checks |
+| routing_strategy_args | dict | None | Routing strategy arguments |
+| routing_groups | list | None | Routing groups |
+| provider_budget_config | dict | None | Provider budget config |
+| alerting_config | dict | None | Alerting config |
+| router_general_settings | dict | None | General router settings |
+| deployment_affinity_ttl_seconds | int | None | Deployment affinity TTL |
+| model_group_affinity_config | dict | None | Model group affinity config |
+| ignore_invalid_deployments | bool | None | Ignore invalid deployments |
+| enable_health_check_routing | bool | None | Enable health check routing |
+| health_check | bool | None | Enable health checks |
+| health_check_timeout | int | None | Health check timeout |
+| health_check_interval | float | None | Health check interval |
+| health_check_max_tries | int | None | Max health check tries |
+| health_check_retry_threshold | int | None | Retry threshold |
+| health_check_retries | int | None | Health check retries |
+| health_check_staleness_threshold | int | None | Staleness threshold |
+| health_check_ignore_transient_errors | bool | None | Ignore transient errors |
+| disable_health_check_on_failing_model | bool | None | Disable health check on failing model |
+| disable_banned_choices | bool | None | Disable banned choices |
+| default_max_retries | int | None | Default max retries |
+| max_parallel_requests | int | None | Max parallel requests |
+| num_limit_per_client | int | None | Num limit per client |
+| max_all_requests | int | None | Max all requests |
+| max_requests | int | None | Max requests |
+| minimum_response_threshold | int | None | Minimum response threshold |
+| enable_token_optimization | bool | None | Enable token optimization |
+| enable_post_routing | bool | None | Enable post routing |
+| enable_pre_routing | bool | None | Enable pre routing |
+| allowed_openai_parameters | list | None | Allowed OpenAI parameters |
+| allowed_url_regex | str | None | Allowed URL regex |
+| allowed_openai_params | list | None | Allowed OpenAI params |
+| contextual_undercut | bool | None | Enable contextual undercut |
+| contextual_undercut_threshold | float | None | Undercut threshold |
+| request_callback_function | str | None | Request callback function |
+| request_fail_callback_function | str | None | Request fail callback function |
+| response_callback_function | str | None | Response callback function |
+| before_appends | list | None | Before append callbacks |
+| before_add_sequence | list | None | Before add sequence callbacks |
+| after_add_sequence | list | None | After add sequence callbacks |
+| before_callback | str | None | Before callback function |
+| after_callbacks | list | None | After callback functions |
+| before_soft_id | str | None | Before soft ID |
+| max_soft_cost_tokens | float | None | Max soft cost tokens |
+| max_soft_budget_tokens | float | None | Max soft budget tokens |
+| max_soft_budget_time | int | None | Max soft budget time |
+| soft_timeout | float | None | Soft timeout |
+| request_timeout | float | None | Request timeout |
+| request_timeout_retry | bool | None | Request timeout retry |
+| max_timeout | float | None | Max timeout |
+| max_retries | int | None | Max retries |
+| llm_response_parsing_attempts_retries | int | None | LLM response parsing retries |
+| max_wait | float | None | Max wait time |
+| enable_health_check_routing | bool | None | Enable health check routing |
+| enable_weighted_failover | bool | None | Enable weighted failover |
+| cooldown_time | float | None | Cooldown time |
+| deployment_affinity_ttl_seconds | int | None | Deployment affinity TTL |
+| ignore_invalid_deployments | bool | None | Ignore invalid deployments |
+| enable_pre_call_checks | bool | None | Enable pre-call checks |
+| routing_strategy | str | None | Routing strategy |
+| model_group_alias | dict | None | Model group aliases |
+
+### Other Settings
+
+Additional router and proxy configuration.