Sync variable-length-quantity with problem-specifications

[senekor]

Notably, this removes any tests about overflow as well as the
overflow variant from the error enum in the stub.
Overflow handling is a valuable thing to learn and practice,
but there are other exercises which do that.
So it seems unnecesary to expand the case only on the Rust track
and maintain our own additional tests.

This is a backwards compatible change,
submissions which choose to handle overflow should still work fine.

[senekor]

Current dependencies on/for this PR:

• main
  • PR Add test generator #1722
    • PR Use todo instead of unimplemented #1723
      • PR Remove test_ prefix from tests #1727
        • PR Add tera filter to_hex for use in test templates #1732
          • PR Symlink problem-specifications for easy reference #1737
            • PR Sync pascals-triangle with problem-specifications #1743
            • PR Sync custom-set with problem-specifications #1741
            • PR doubly-linked-list: Fix inaccurate instructions #1740
            • PR Sync two-bucket with problem-specifications #1739
            • PR Sync variable-length-quantity with problem-specifications #1731 👈
            • PR Sync word-count with problem-specifications #1729
        • PR Sync wordy with problem-specifications #1728
        • PR Decimal: Compare different numbers of digits #1726
        • PR Improve test order of low-power-embedded-game #1725
        • PR Improve example solution of circular-buffer #1724

This comment was auto-generated by Graphite.

[senekor]

It looks like the test suite here was significantly out of sync with problem-specifications. I should go through the existing tests and see which of them are worth preserving / upstreaming.

[senekor]

Apart from the overflow thing, I did not investigate much further what the overlap of the two test suites is. They are very different, trying to figure out a submission that passes one test suite but not the other would've been difficult.