feat(catalog): replace Databricks model-label tokenizer with models.dev registry lookup

The frontend heuristic in #3586 title-cased every word and converted adjacent
digit pairs to decimals for any string starting with 'databricks-'. This
invented false labels for custom workspace endpoints: 'databricks-team-2025-01'
→ 'Team 2025.01', 'databricks-finance-2025-01-30' → 'Finance 2025.01 30'.

Replace with a lookup-and-pass-through approach drawn from the same design
goose uses in production:

- Generator script (scripts/generate-databricks-model-names.py) fetches
  https://models.dev/api.json and emits a sorted Rust static slice of
  (id, name) pairs under crates/buzz-agent/src/databricks_model_names.rs.
  Refresh by rerunning the script and committing the diff.

- catalog.rs consults the table via databricks_model_name(id): known managed
  endpoints get curated names (e.g. 'databricks-gpt-5-5' → 'GPT-5.5'),
  unknown/custom endpoints return their raw ID unchanged. Applied to all
  ModelEntry construction paths: v1/v2 discovery, the empty-list fallback,
  and discovery_failure_fallback.

- Frontend: a matching TS registry (desktop/src/features/agents/lib/
  databricksModelNames.ts) covers persisted raw IDs on cards, rows, and
  popovers that render before discovery data is available. formatAgentModelLabel
  and formatDefaultModelLabel use the registry; no heuristic string mangling
  exists anywhere in the TS layer.

- AGENTS.md documents the three-tier label precedence: API/runtime name first,
  table-backed fallback second, raw ID last.

Supersedes #3586 (kennylopez-model-display-labels).

Co-authored-by: Will Pfleger <pfleger.will@gmail.com>
Signed-off-by: Will Pfleger <pfleger.will@gmail.com>

scripts/generate-databricks-model-names.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python3
+"""Generate crates/buzz-agent/src/databricks_model_names.rs from models.dev.
+
+Usage:
+    python3 scripts/generate-databricks-model-names.py \
+        > crates/buzz-agent/src/databricks_model_names.rs
+
+Fetches https://models.dev/api.json, extracts providers.databricks.models (the
+authoritative curated name registry used by goose and others), and emits a
+static Rust slice of (id, display_name) pairs sorted by ID.
+
+Re-run whenever Databricks ships a new managed endpoint and commit the diff.
+"""
+
+import json
+import subprocess
+import sys
+
+
+URL = "https://models.dev/api.json"
+
+
+def fetch(url: str) -> bytes:
+    # urllib blocks with HTTP 403 without a browser User-Agent; use curl when
+    # available so the script works in hermit environments where curl is pinned.
+    result = subprocess.run(
+        ["curl", "-s", "-A", "Mozilla/5.0", url],
+        capture_output=True,
+        check=True,
+    )
+    return result.stdout
+
+
+def rust_str(s: str) -> str:
+    """Emit a Rust string literal with double quotes."""
+    escaped = s.replace("\\", "\\\\").replace('"', '\\"')
+    return f'"{escaped}"'
+
+
+def main() -> None:
+    raw = fetch(URL)
+    data = json.loads(raw)
+    models: dict = data["databricks"]["models"]
+    entries = sorted(
+        (k, v["name"] if isinstance(v, dict) else k) for k, v in models.items()
+    )
+
+    lines = [
+        "// GENERATED by scripts/generate-databricks-model-names.py",
+        "// Source:  https://models.dev/api.json -- providers.databricks.models",
+        "// Refresh: python3 scripts/generate-databricks-model-names.py \\",
+        "//              > crates/buzz-agent/src/databricks_model_names.rs",
+        "//",
+        "// Do not hand-edit -- rerun the script to update.",
+        "",
+        "/// Curated display names for known Databricks AI Gateway endpoints.",
+        "///",
+        "/// Keys are endpoint IDs returned verbatim by the discovery APIs.",
+        "/// Values are human-readable display names sourced from models.dev.",
+        "///",
+        "/// Unknown endpoint IDs are displayed as their raw ID -- no guessing.",
+        "pub(crate) static DATABRICKS_MODEL_NAMES: &[(&str, &str)] = &[",
+    ]
+    for id_, name in entries:
+        lines.append(f"    ({rust_str(id_)}, {rust_str(name)}),")
+    lines += ["];", ""]
+
+    sys.stdout.write("\n".join(lines))
+
+
+if __name__ == "__main__":
+    main()