Simplify FixedLengthEncoding

[tustvold]

Which issue does this PR close?

Part of #2677

Rationale for this change

Const generics have a lot of rough edges, including rust-lang/rust#80528

They were always a bit overkill for the needs here, it was only a lack of imagination on my part that made me reach for them 😅

What changes are included in this PR?

Are there any user-facing changes?

No, this trait is an implementation detail of the row format

[alamb]

makes sense to me -- do we need to run any benchmarks?

[tustvold]

Benchmarks show no major change in performance, this doesn't change any of the types involved, just the trait plumbing, and so it would be surprising if this changed performance (although stranger things have happened)

[ursabot]

Benchmark runs are scheduled for baseline = 70054fd and contender = 931c6fc. 931c6fc is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ec2-t3-xlarge-us-east-2] ec2-t3-xlarge-us-east-2
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on test-mac-arm] test-mac-arm
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ursa-i9-9960x] ursa-i9-9960x
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ursa-thinkcentre-m75q] ursa-thinkcentre-m75q
Buildkite builds:
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python, R. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java