chore: compute artifact digests by reading files in chunks

[Copilot]

Hashes.from_path was loading entire artifact files into memory via path.read_bytes() before computing SHA1/SHA256 digests — a problem for large artifacts.

Changes

• craft_application/models/manifest.py: Replaced path.read_bytes() with incremental 64KB chunk reads, feeding each chunk to both hash objects simultaneously.

# Before
read_bytes = path.read_bytes()
sha1=hashlib.sha1(read_bytes).hexdigest(),
sha256=hashlib.sha256(read_bytes).hexdigest(),

# After
with path.open("rb") as f:
    while chunk := f.read(65536):
        sha1.update(chunk)
        sha256.update(chunk)

This matches the existing chunked-read pattern already used in remote/utils.py. Once Python 3.10 support is dropped, this can be replaced with hashlib.file_digest().

[Copilot]

Pull request overview

This PR updates artifact digest computation to avoid loading entire files into memory, improving behavior for large artifacts.

Changes:

• Reworked Hashes.from_path() to stream file contents in 64KB chunks instead of path.read_bytes()
• Updated digest computation to reuse hashlib objects via incremental update() calls

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.