feat: function-calling (MCP-style) tests plus ultra-hard tasks

Adds real OpenAI function-calling support end to end and two test tranches.

Client (internal/llm): Request.Tools advertises tools with tool_choice
auto; Response.ToolCalls carries the model's calls, decoding each
function's JSON-encoded argument string into an object (malformed
arguments preserved raw with Decoded=false). Verified live: the gateway
round-trips tools and the qwen/deepseek tool-call parsers emit the
expected format.

Runner: when a test declares Tools, it serializes the model's tool calls
into a canonical JSON envelope and scores that, not the free text.

Eval (toolcall.go): ToolCalled, NoToolCalled, ToolCallWithArgs (subset
match, number/string coercion), ToolSequence, FirstToolIs, ToolNotCalled,
ToolArgValuesCover - all parse the envelope, no llm dependency.

Tests: 12 function-calling tests (tools/function-calling) covering tool
selection, argument extraction, enum constraints, chained and parallel
calls, no-tool-needed and missing-param abstention, destructive-tool
safety, and name-vs-description traps; 8 ultra-hard single-shot tasks in
existing subcategories. Ground-truth self-consistency tests assert every
documented answer and ideal tool envelope scores 1.0.

internal/tests/advanced_groundtruth_test.go

@@ -27,6 +27,14 @@ func TestAdvancedGroundTruthScoresFull(t *testing.T) {
 		"adv-code-conflicting-evidence":      `{"port": 9090, "authoritative": "config"}`,
 		"adv-sec-exploit-chain":              `{"vuln": "open-redirect"}`,
 		"adv-sec-rotation-blast-radius":      `30`,
+		"ultra-c-int8-signed":                `-76`,
+		"ultra-hard-modexp":                  `5`,
+		"ultra-hard-fp-equality":             `{"equal": false}`,
+		"ultra-py-bankers-round":             `6`,
+		"ultra-sec-redos":                    `{"vulnerable": true}`,
+		"ultra-scen-cap-partition":           `{"sacrificed": "availability"}`,
+		"ultra-ai-pq-bytes":                  `16`,
+		"ultra-sql-deadlock-cycle":           `{"fix": "consistent-lock-order"}`,
 	}
 
 	byID := map[string]int{}