Custom connections are gone: the CLI talks to the LLeMbas server it is logged in to

LLeMbas CLI 2.0 is LLeMbas-server-specific (CLI rule 2). Removed: the
anthropic, gemini, ollama and responses clients, context discovery and
llama-swap unloading (discover.ts), the Connection schema with its dialect,
auth, key_cmd, discover, unload_url and one_model_at_a_time, the model
`fallback` (it only ever moved to another server), effort_map and cache, and
the multi-connection refs. Kept: openai-chat, http, sse, think, effort,
tokens and all of src/lembas.

- connections.yaml holds the login (`type: webui`, as lembas login writes
  it). A custom entry from 1.x is read past with a warning that says how to
  log in, so an old file still starts; with several logins left, the latest
  is used and the others named.
- A model's ref is the id the server serves it under; the older forms
  (`<login>/<id>`, `<login>/<model>`) still resolve and config.yaml is
  rewritten once. Without a login there is no model, and the error (and the
  first-run screen, now the sign-in alone) says `lembas login <address>`.
- A login to another server signs this machine out of the previous one.
- The window is the server's word (no discovery); a capacity group comes
  from the server's `capacity.group` alone; embeddings use the server's
  /v1/embeddings.
- `lembas models` lists the server's models; install.sh's starter files, the
  README and the JSON schemas say the same.

Tests: every fixture is a login now (the fake server lists models, answers
other GETs with 404, and can hand one model to another script, as a server
with several backends); the dialect suites and the cross-connection tests are
gone with their code.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

src/tui/components/setup.tsx

@@ -1,18 +1,9 @@
 import { TextAttributes } from "@opentui/core"
 import { useKeyboard } from "@opentui/solid"
-import { join } from "node:path"
 import { createSignal, Show } from "solid-js"
-import { paths } from "../../config/paths.ts"
-import { ModelError } from "../../provider/index.ts"
+import { ModelError, NOT_LOGGED_IN } from "../../provider/index.ts"
 import { useTui } from "../context.ts"
 
-const EXAMPLE = `connections:
-  local:                          # any name; llama.cpp, LM Studio, vLLM, Ollama's /v1…
-    dialect: openai-chat
-    base_url: http://127.0.0.1:8080/v1
-    models:
-      qwen3: { context: 131072 }`
-
 /** Before there is a session: no model to talk to (a fresh install), or a config file that cannot
  *  be read. Says what to do and where; r tries again once the files are fixed, q or ctrl+c leaves.
  *  Not an exception: thrown with the terminal still the TUI's, mouse reporting would stay on, and
@@ -38,7 +29,7 @@ export function Setup(p: { error: Error; onRetry: () => void; onQuit: () => void
     try {
       const { codeInstructions, login } = await import("../../lembas/login.ts")
       const r = await login(target, { onCode: (start, d) => setStatus(codeInstructions(start, d)) })
-      setStatus(`Signed in: ${r.models.length} model${r.models.length === 1 ? "" : "s"} as ${r.connection}/…${r.instanceDefault ? `, starting on ${r.connection}/${r.instanceDefault}` : ""}.`)
+      setStatus(`Signed in: ${r.models.length} model${r.models.length === 1 ? "" : "s"}${r.instanceDefault ? `, starting on ${r.instanceDefault}` : ""}.`)
       p.onRetry()
     } catch (e) {
       setStatus((e as Error).message)
@@ -65,11 +56,11 @@ export function Setup(p: { error: Error; onRetry: () => void; onQuit: () => void
   return (
     <box flexDirection="column" width="100%" height="100%" paddingLeft={2} paddingTop={1} paddingRight={2} gap={1}>
       <text fg={c().title} attributes={TextAttributes.BOLD}>
-        {theme().branding.agentName} — {noModel() ? "no model to talk to yet" : "the configuration has a mistake"}
+        {theme().branding.agentName} — {p.error.message === NOT_LOGGED_IN ? "not logged in to a LLeMbas server" : noModel() ? "no model to talk to" : "the configuration has a mistake"}
       </text>
-      {/* "No model configured" is what the rest of the screen explains; anything else (a model
-          that does not exist, a file that does not parse) is shown as it is. */}
-      <Show when={!p.error.message.startsWith("No model configured")}>
+      {/* Not being logged in is what the rest of the screen explains; anything else (a model the
+          server does not serve, a file that does not parse) is shown as it is. */}
+      <Show when={p.error.message !== NOT_LOGGED_IN}>
         <text fg={noModel() ? c().text : c().error} wrapMode="word">
           {p.error.message}
         </text>
@@ -80,7 +71,7 @@ export function Setup(p: { error: Error; onRetry: () => void; onQuit: () => void
           when={typing()}
           fallback={
             <text fg={c().muted} wrapMode="word">
-              {"Sign in to a LLeMbas instance and use its models: press l (or, from a shell, lembas login <address>)."}
+              {"LLeMbas CLI uses the models of the LLeMbas server it is logged in to. Log in: press l (or, from a shell, lembas login <address>)."}
             </text>
           }
         >
@@ -96,24 +87,9 @@ export function Setup(p: { error: Error; onRetry: () => void; onQuit: () => void
             {status()}
           </text>
         </Show>
-        <text fg={c().muted} wrapMode="word">
-          {`Or, without an instance — nothing is predefined: add the server you use to ${join(paths.config, "connections.yaml")}, for example`}
-        </text>
-        <box paddingLeft={2}>
-          <text fg={c().text}>{EXAMPLE}</text>
-        </box>
-        <text fg={c().muted} wrapMode="word">
-          {`and pick the model in ${join(paths.config, "config.yaml")}:`}
-        </text>
-        <box paddingLeft={2}>
-          <text fg={c().text}>model: local/qwen3</text>
-        </box>
-        <text fg={c().muted} wrapMode="word">
-          A key is never pasted in: api_key: "{"{env:NAME}"}" or "{"{file:~/path}"}". https://llembas.eu/docs/ has the other dialects (Anthropic, OpenAI, Gemini, Ollama).
-        </text>
       </Show>
       <text fg={c().muted} wrapMode="word">
-        From a shell, lembas config check reads the files and says what is wrong, and lembas models lists the models it found.
+        From a shell, lembas config check reads the files and says what is wrong, and lembas models lists the server's models.
       </text>
       <text fg={c().label}>{typing() ? "enter sign in · esc back · ctrl+r try again · ctrl+c quit" : `${noModel() ? "l log in · " : ""}r try again · q quit`}</text>
     </box>