Improving `Largest Contentful Paint` time?

Hi there, I'm using this library for https://www.timecomplexity.ai, and it's working nicely, so thank you for that!

One issue I'm seeing though is that the Largest Contentful Paint time is quite high because of this library:

Curious if you have any suggestions on how to improve this? I'm using it in a next.js site hosted on Vercel like this:

import CodeMirror from '@uiw/react-codemirror'
import {go} from '@codemirror/legacy-modes/mode/go'
import {xcodeDarkInit} from '@uiw/codemirror-theme-xcode'

export default function Analysis() {
  return (
      <CodeMirror
            value={inputCode}
            placeholder={'Use AI to analyze your code\'s runtime complexity.\nWorks across languages and without requiring headers, imports, or a main statement - just the code itself.\nPlease note that results may be incorrect in subtle (or not so subtle) ways, so use at your own risk!'}
            autoFocus
            editable={!loading}
            minHeight="300px"
            extensions={[
              StreamLanguage.define(go),
              EditorView.lineWrapping
            ]}
            theme={xcodeDarkInit({
              settings: {
                fontFamily: inputCode.length === 0 ? 'var(--font-inter)' : ''
              }
            })}
            onChange={(value: any) => {
              if (value.length > MAX_CODE_LENGTH) {
                setErrorMessage(`Please enter code less than ${MAX_CODE_LENGTH} characters. You are currently at ${value.length} characters.`)
              } else if (typeof errorMessage !== 'string' || typeof errorMessage === 'string' && errorMessage.length !== 0) {
                setErrorMessage('')
              }
              setInputCode(value)
            }}
          />
    )
 }

Start with the CodeMirror usage shown in the issue and profile its impact on Largest Contentful Paint in the reported Next.js and Vercel setup. Compare the editor's loading and rendering behavior with the page's LCP measurement; done means identifying a reproducible cause and a clearly validated improvement.