panic/unreachable from udp.data function in benchmarks using udp4 proptest strategy

## Describe the bug?
Using the `udp4` proptest strategy to generate UDP4 packets in bench mark test, I noticed when making a call to that packet's `data()` function I'd get an unreachable panic. In adding some instrumentation, it is panicking because the payload_offset is 42 and the data buffer length itself is also 42. I did notice in adding println to the function that the issue would go away (memory/concurrency issue maybe?). I don't see it with my own strategy that uses the udp4 call but then extends the packet with a payload, which is doesn't have in the proptest generation. But then the payload may be masking it. Will know if that's true with some more testing.

I made a [slight change](https://github.com/sbuzzard/capsule/blob/topic/data-offset-err-with-udp4-strategy/core/src/packets/udp.rs#L153-L164) to udp4 in the `data()` function as part of this repoduction to get the specific error.

## Steps to reproduce?
[this branch in my fork](https://github.com/sbuzzard/capsule/blob/topic/data-offset-err-with-udp4-strategy) demonstrates it with [this change](https://github.com/sbuzzard/capsule/blob/topic/data-offset-err-with-udp4-strategy/bench/packets.rs#L37)

### Capsule version?
I am using the `djin/borrow-mut-parse` branch (0.0.2 pre-alpha) - my branch above was made from that.

### OS?
Using the capsule sandbox

### Docker / VM / Bare?
Using the capsule sandbox with vagrant but not with the docker image in vagrant.

### Stack trace or error log output
```
Benchmarking packets::single_peek_vs_parse_on_udp/packets::single_parse_udp: Warming up for 3.0000 sError in data buffer: Offset 42 exceeds the buffer length 42.
thread 'main' panicked at 'internal error: entered unreachable code', /opt/sources/dpdk/capsule/capsule/core/src/packets/udp.rs:161:17
stack backtrace:
0: rust_begin_unwind
at /rustc/9d1b2106e23b1abd32fce1f17267604a5102f57a/library/std/src/panicking.rs:498:5
1: core::panicking::panic_fmt
at /rustc/9d1b2106e23b1abd32fce1f17267604a5102f57a/library/core/src/panicking.rs:116:14
2: core::panicking::panic
at /rustc/9d1b2106e23b1abd32fce1f17267604a5102f57a/library/core/src/panicking.rs:48:5
3: as core::iter::traits::iterator::Iterator>::fold
4: criterion::bencher::Bencher::iter_custom
5: as criterion::routine::Routine>::warm_up
6: criterion::routine::Routine::sample
7: criterion::analysis::common
8: criterion::benchmark_group::BenchmarkGroup::bench_function
9: packets::main
```

Start with core/src/packets/udp.rs around the udp4 data() implementation and compare it with the reproduction change linked in the issue. Run the benchmark in bench/packets.rs using the udp4 proptest strategy, then check the payload_offset and data-buffer behavior shown by the stack trace. Done means the benchmark no longer reaches the unreachable panic for generated UDP4 packets.

Written by the indexing model from the issue text.