test: add eval test setup W-19811634 W-19812303

[cristiand391]

What does this PR do?

Adds:

Discoverability tests

Adds tool prediction scorer based on Sentry's one with some updates:

• updated sys prompt to avoid over-indexing on eval expectations (was causing too much false positives with score 1)
• enforce expected tools are available to influence final scoring (will be blocked in code later)
• scorer updated to send the full tool metadata (Sentry's one sends tool name + 1st description line) to enhance scoring based on parameters and examples availables

test failure example:

E2E eval tests

Do a full agent loop with the DX MCP tools, see the TESTING.md for more examples and the packages/mcp/test/evals/e2e/ tests.

Others:

• updated the GHA workflow to only run on linux (E2E tool tests cover cross OS compat, eval tests aren't OS-dependant)
• split eval tests in 2 types, each has its own vitest config/pjson script and run in separate CI jobs
• Set up Gemini API key in repo

What issues does this PR fix or reference?

@W-19811634@
@W-19812303@