feat: add from-readme command for README-driven GIF generation

[israkir]

Summary

• add a new vhs from-readme command that extracts shell commands from Markdown fenced blocks and records them into a GIF
• support focused extraction with --section and --command, plus safer execution controls (--wait-timeout, --wait-pattern) and a --dry-run/--tape-out workflow
• register the new command in the root CLI and document usage, flags, and examples in README.md
• add unit tests for README discovery, Markdown parsing, command filtering, duration formatting, and tape generation

Why

Creating demos from documentation currently requires manually rewriting commands into a tape. This feature turns README command examples into a reproducible recording flow, reducing copy/paste overhead and helping keep docs and demos aligned.

PS. I just needed this feature for my own project and thought it would be cool to share with the users of this project :)

Implementation Details

• introduces fromreadme.go with:
  • CLI option parsing (fromReadmeOptions)
  • README auto-discovery (findREADME)
  • Markdown shell command extraction (extractShellCommands)
  • optional command filtering (filterByCommand)
  • tape generation (buildFromReadmeTape)
• wires fromReadmeCmd into main.go
• extends README.md with a dedicated From README section and full flag table
• adds fromreadme_test.go to cover core behavior and edge cases

Test Plan

• go test ./...
• vhs from-readme README.md --dry-run
• vhs from-readme README.md --command copinance --dry-run
• vhs from-readme README.md --section "Using the CLI" --dry-run
• vhs from-readme README.md --command copinance --output demo.gif and verify GIF is created