precompiles: Handle trivial inputs to expmod

[chfast]

Before looking up an answer for the expmod precompile in stub,
check if the input has at least the bytes encoding the value lengths.
If not or if the modulus length is 0 then the answer is trivially
the sequence of zero bytes of the length of the modulus.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.54%. Comparing base (d84b471) to head (209afb2).
Report is 11 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1163      +/-   ##
==========================================
- Coverage   94.54%   94.54%   -0.01%     
==========================================
  Files         168      168              
  Lines       18254    18245       -9     
==========================================
- Hits        17259    17250       -9     
  Misses        995      995              

Flag	Coverage Δ
eof_execution_spec_tests	20.55% <80.00%> (-1.92%)	⬇️
ethereum_tests	28.25% <100.00%> (-0.04%)	⬇️
ethereum_tests_silkpre	20.99% <ø> (+0.01%)	⬆️
execution_spec_tests	20.92% <100.00%> (-0.04%)	⬇️
unittests	89.86% <0.00%> (+0.04%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
test/state/precompiles.cpp	98.98% <100.00%> (+0.01%)	⬆️
test/state/precompiles_stubs.cpp	99.41% <ø> (-0.03%)	⬇️

🚀 New features to boost your workflow:

• ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[rodiazet]

Looks good. As a comment I would also add less trivial check. In case when input size is less or equal than 96+base_len+exp_len it such a case modulus is also 0. So the result is 0 too. This can be added in a separated PR.

[chfast]

Looks good. As a comment I would also add less trivial check. In case when input size is less or equal than 96+base_len+exp_len it such a case modulus is also 0. So the result is 0 too. This can be added in a separated PR.

This is correct, but I'm not sure we want to go this path. I.e. we don't want do handle degenerated cases separately unless necessary.