Automatically discover and configure Bifrost-backed LLM providers

[crspeller]

Summary

This PR updates the plugin to derive supported LLM backends from Bifrost instead of a plugin-maintained allowlist.

Key changes:

• replaced hardcoded service-type mapping and validation with dynamic support based on Bifrost schemas.StandardProviders
• kept a simple openaicompatible convenience option while adding advanced Bifrost passthrough fields for provider-specific setup:
  • bifrostKeyJSON
  • bifrostProviderConfigJSON
• refactored runtime setup, model fetching, and transcription setup to use a shared service-to-Bifrost resolver
• changed model fetching to use the full service config instead of only serviceType, apiKey, apiURL, and orgID
• normalized fetched model IDs using Bifrost parsing semantics instead of naive slash stripping
• made the system console provider dropdown dynamic and added advanced Bifrost JSON inputs
• removed stale provider plumbing and fully removed asage references
• updated docs and test coverage, including UI coverage for dynamic providers and advanced config persistence
• fixed a follow-up CI lint regression in the system console service editor

Validation performed:

• cd /workspace/webapp && npm run check-types
• docker run --rm -v /workspace:/workspace -w /workspace golang:tip go test ./llm/... ./bifrost/...
• docker run --rm -v /workspace:/workspace -w /workspace golang:tip go test ./api/... ./server/...
• cd /workspace/e2e && npx playwright test tests/system-console/service-management-ui.spec.ts --reporter=list

Ticket Link

NONE

Screenshots

NONE

Release Note

Updated Mattermost Agents to discover supported Bifrost-backed LLM providers dynamically, added advanced Bifrost provider configuration fields in the system console, removed asksage references, and updated model fetching to use full service configuration with Bifrost-aware model normalization.

  

[cursor]

Cursor Agent can help with this pull request. Just @cursor in comments and I'll start working on changes in this branch.
_{Learn more about Cursor Agents}

[github-actions]

🤖 LLM Evaluation Results

OpenAI

⚠️ Overall: 18/19 tests passed (94.7%)

Provider	Total	Passed	Failed	Pass Rate
⚠️ OPENAI	19	18	1	94.7%

❌ Failed Evaluations

Show 1 failures

OPENAI

1. TestReactEval/[openai]_react_cat_message

• Score: 0.00
• Rubric: The word/emoji is a cat emoji or a heart/love emoji
• Reason: The output is the literal text "heart_eyes_cat", not an actual cat emoji (e.g., 😺/🐱) or a heart/love emoji (e.g., ❤️/😍).

---

Anthropic

✅ Overall: 19/19 tests passed (100.0%)

Provider	Total	Passed	Failed	Pass Rate
✅ ANTHROPIC	19	19	0	100.0%

---

This comment was automatically generated by the eval CI pipeline.

[mm-cloud-bot]

Creating a Plugin SpinWick test server

[mm-cloud-bot]

Plugin Spinwick PR #569 🎉

Test server created!

Access here: https://agents-pr-569-auqk4.test.mattermost.cloud

Plugin	Version	Artifact
agents	c17dafb	Download

Installation ID: ko6kci4apfgaxnaj8a7f4c1gxe
Logs: Click here

Credentials: Posted securely in this Mattermost channel - Look for PR #569