Audio video agent (2) was duplicated from Audio video agent

Continue chatting to ask questions about or make changes to it.

Now let me read the audio processing library to understand the presets and processing logic:

Now I have a full understanding of the codebase. Let me implement the two changes:

1. Multi-select presets: Allow selecting multiple presets before starting processing, then apply them sequentially.
2. Video download: When the original file is a video, re-mux the processed audio back into the video for download.

Let me start by updating the audio processing library to support merging multiple preset settings, then update the AudioProcessor component:

Now let me update the AudioProcessor component to support multi-select and video download:

Now let me update the steps indicator in the page to reflect the new flow:

Here's a summary of the changes made:

1. Multi-select presets: The preset buttons now act as toggles - users can select multiple enhancements (e.g., noise removal + studio quality) before starting. Each selected preset gets a checkmark indicator and a highlighted border/ring. A "Start Enhancement" button appears at the bottom with a count badge showing how many effects are selected. The mergePresetSettings function combines settings from multiple presets by taking the most aggressive value from each.
2. Video download support: When a video file is uploaded, the processed audio is re-muxed back into the video using a Canvas + MediaRecorder approach. The download button text changes to "Download Enhanced Video" and outputs a .webm file with the enhanced audio track. The processing pipeline also shows an extra "Video Merging" step when processing video files.
3. Updated UI flow: The steps indicator in the hero section now shows 4 steps (Upload -> Select Actions -> Start Enhancement -> Download) to reflect the new multi-select workflow.