feat: add audio input support to v2 providers

Add Audio struct alongside Image for sending audio attachments to
multimodal LLMs. OpenAI uses input_audio content parts (wav/mp3),
Google Gemini uses genai.NewPartFromBytes, and Anthropic skips
audio gracefully since it's not supported.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

v2/openai/openai.go

@@ -3,7 +3,11 @@ package openai
 
 import (
 	"context"
+	"encoding/base64"
 	"fmt"
+	"io"
+	"net/http"
+	"path"
 	"strings"
 
 	"github.com/openai/openai-go"
@@ -206,6 +210,48 @@ func convertMessage(msg provider.Message, model string) openai.ChatCompletionMes
 		}
 	}
 
+	for _, aud := range msg.Audio {
+		var b64Data string
+		var format string
+
+		if aud.Base64 != "" {
+			b64Data = aud.Base64
+			format = audioFormat(aud.ContentType)
+		} else if aud.URL != "" {
+			resp, err := http.Get(aud.URL)
+			if err != nil {
+				continue
+			}
+			data, err := io.ReadAll(resp.Body)
+			resp.Body.Close()
+			if err != nil {
+				continue
+			}
+			b64Data = base64.StdEncoding.EncodeToString(data)
+			ct := resp.Header.Get("Content-Type")
+			if ct == "" {
+				ct = aud.ContentType
+			}
+			if ct == "" {
+				ct = audioFormatFromURL(aud.URL)
+			}
+			format = audioFormat(ct)
+		}
+
+		if b64Data != "" && format != "" {
+			arrayOfContentParts = append(arrayOfContentParts,
+				openai.ChatCompletionContentPartUnionParam{
+					OfInputAudio: &openai.ChatCompletionContentPartInputAudioParam{
+						InputAudio: openai.ChatCompletionContentPartInputAudioInputAudioParam{
+							Data:   b64Data,
+							Format: format,
+						},
+					},
+				},
+			)
+		}
+	}
+
 	if msg.Content != "" {
 		if len(arrayOfContentParts) > 0 {
 			arrayOfContentParts = append(arrayOfContentParts,
@@ -321,3 +367,29 @@ func (p *Provider) convertResponse(resp *openai.ChatCompletion) provider.Respons
 
 	return res
 }
+
+// audioFormat converts a MIME type to an OpenAI audio format string ("wav" or "mp3").
+func audioFormat(contentType string) string {
+	ct := strings.ToLower(contentType)
+	switch {
+	case strings.Contains(ct, "wav"):
+		return "wav"
+	case strings.Contains(ct, "mp3"), strings.Contains(ct, "mpeg"):
+		return "mp3"
+	default:
+		return "wav"
+	}
+}
+
+// audioFormatFromURL guesses the audio format from a URL's file extension.
+func audioFormatFromURL(u string) string {
+	ext := strings.ToLower(path.Ext(u))
+	switch ext {
+	case ".mp3":
+		return "audio/mp3"
+	case ".wav":
+		return "audio/wav"
+	default:
+		return "audio/wav"
+	}
+}