fix(server): count quota by the key actually used, and more review fixes

Found by the PR review, each with a test that failed first:
- Quota: any key header skipped it, even one the provider never reads
  (x-aws-access-key-id with OpenAI), so a request ran on the server's
  key without being counted. The check now runs after the model is
  resolved and uses usesServerCredentials. On main already.
- usesServerCredentials read the raw base URL; "/" cleans up to none, so
  an Ollama request ran on the server's key past the server-model check.
- SGLang's default 127.0.0.1:8000 only fills the settings form. Chat and
  the model list used it as a real address, so the server called its own
  machine even with private URLs blocked. Now a base URL is required.
- With a user's OpenAI key and no base URL, the SDK read the server's
  OPENAI_BASE_URL. The official endpoint is now passed. On main already.
- The Test button refused nothing on the server's keys (Ollama Cloud),
  and a 15 s timeout reported "connected, no tool call".
- The model list for Ollama without a base URL came from ollama.com while
  chat went to the server's Ollama.
- Bedrock's "Too many tokens, please wait" counted as context too long.
- On the server's keys the provider's error text stays in the server log;
  it can name the server's AWS account, role or internal hosts.
- Desktop app: the preset keys are the user's own (NEXT_AI_DRAWIO_DESKTOP),
  so Max Output Tokens can be raised and keyless models in settings work
  again. A launch that found the remembered port taken no longer replaces
  it, which hid the user's chats and settings for good.

lib/llm-errors.ts

@@ -36,7 +36,8 @@ export interface LLMError {
 // error can come as 403 or 429, a context or image error as a plain 400
 const SPECIFIC_TEXTS: Array<[RegExp, LLMErrorCode]> = [
     [
-        /context length|context window|maximum context|prompt is too long|input is too long|too many (?:input )?tokens/i,
+        // Not "too many tokens": that is Bedrock's throttling message
+        /context length|context window|maximum context|prompt is too long|input is too long|too many input tokens/i,
         "context_too_long",
     ],
     [
@@ -110,12 +111,19 @@ function problemDetail(body: string): string | undefined {
 /**
  * The error text for the chat stream: what went wrong with the provider as
  * JSON for the hint, or the text the model must read to fix a tool call.
+ * On the server's keys the provider's own text stays in the server log:
+ * it can name the server's account, role or internal hosts.
  */
-export function streamErrorText(error: unknown): string {
+export function streamErrorText(error: unknown, hideDetails = false): string {
     // The SDK passes an invalid tool call's error as a plain string
     if (typeof error === "string") return error
     if (isToolCallError(error)) return (error as Error).message
-    return JSON.stringify(classifyLLMError(error))
+    const classified = classifyLLMError(error)
+    if (hideDetails) {
+        console.error("[chat] Provider error:", error)
+        classified.message = "The provider returned an error."
+    }
+    return JSON.stringify(classified)
 }
 
 /**