feat: load models through provider adapters

Provider model selection had outgrown a single hardcoded service.

The old service mixed shared caching with provider catalogs and CLI lookup details.

That made stale model lists more likely as providers changed on separate schedules.

Move model discovery behind each provider so lookup lives next to the integration.

The shared service now focuses on provider resolution, caching, persistence, and dedupe.

Return cache metadata and add bypassCache because model availability changes outside the app.

The UI and /models command can show freshness and let users force a provider refresh.

Surface model descriptions while keeping fallback catalogs for unavailable CLIs or SDKs.

server/modules/providers/list/codex/codex-models.provider.ts

@@ -0,0 +1,88 @@
+import { readFile } from 'node:fs/promises';
+import os from 'node:os';
+import path from 'node:path';
+
+import type { IProviderModels } from '@/shared/interfaces.js';
+import type { ProviderModelOption, ProviderModelsDefinition } from '@/shared/types.js';
+import { readObjectRecord, readOptionalString } from '@/shared/utils.js';
+
+export const CODEX_FALLBACK_MODELS: ProviderModelsDefinition = {
+  OPTIONS: [
+    { value: 'gpt-5.5', label: 'gpt-5.5' },
+    { value: 'gpt-5.4', label: 'gpt-5.4' },
+    { value: 'gpt-5.4-mini', label: 'gpt-5.4-mini' },
+    { value: 'gpt-5.3-codex', label: 'gpt-5.3-codex' },
+    { value: 'gpt-5.2', label: 'gpt-5.2' },
+  ],
+  DEFAULT: 'gpt-5.4',
+};
+
+type CodexCachedModel = {
+  slug?: string;
+  display_name?: string;
+  description?: string;
+  priority?: number;
+  visibility?: string;
+  supported_in_api?: boolean;
+};
+
+const CODEX_MODELS_CACHE_PATH = path.join(os.homedir(), '.codex', 'models_cache.json');
+
+const isCodexCachedModel = (value: unknown): value is CodexCachedModel => {
+  const record = readObjectRecord(value);
+  return Boolean(record && readOptionalString(record.slug));
+};
+
+const readCodexPriority = (value: unknown): number => (
+  typeof value === 'number' && Number.isFinite(value) ? value : Number.MAX_SAFE_INTEGER
+);
+
+const mapCodexModel = (model: CodexCachedModel): ProviderModelOption => ({
+  value: model.slug as string,
+  label: readOptionalString(model.display_name) ?? (model.slug as string),
+  description: readOptionalString(model.description),
+});
+
+const buildCodexModelsDefinition = (models: CodexCachedModel[]): ProviderModelsDefinition => {
+  const sortedModels = [...models]
+    .filter((model) => model.visibility !== 'hidden' && model.supported_in_api !== false)
+    .sort((left, right) => readCodexPriority(left.priority) - readCodexPriority(right.priority));
+
+  const options: ProviderModelOption[] = [];
+  const seenValues = new Set<string>();
+
+  for (const model of sortedModels) {
+    const mappedModel = mapCodexModel(model);
+    if (seenValues.has(mappedModel.value)) {
+      continue;
+    }
+
+    seenValues.add(mappedModel.value);
+    options.push(mappedModel);
+  }
+
+  if (options.length === 0) {
+    return CODEX_FALLBACK_MODELS;
+  }
+
+  return {
+    OPTIONS: options,
+    DEFAULT: options[0]?.value ?? CODEX_FALLBACK_MODELS.DEFAULT,
+  };
+};
+
+export class CodexProviderModels implements IProviderModels {
+  async getSupportedModels(): Promise<ProviderModelsDefinition> {
+    try {
+      const raw = await readFile(CODEX_MODELS_CACHE_PATH, 'utf8');
+      const parsed = readObjectRecord(JSON.parse(raw));
+      const models = Array.isArray(parsed?.models)
+        ? parsed.models.filter(isCodexCachedModel)
+        : [];
+
+      return buildCodexModelsDefinition(models);
+    } catch {
+      return CODEX_FALLBACK_MODELS;
+    }
+  }
+}