Improve Conv1D layers (Eigen backend)

[stepanmk]

Changes made:

• conv layers now use a circular buffer instead of a double buffered state
• addition of setStatePointers() method for efficient selection of particular buffer positions
• addition of state_ptrs variable to store the buffer positions
• addition of state_cols variable to store the selected buffer columns for convolution
• changed size of some of the Conv1D and Conv1DT member variables due to different state size
• edited Conv1D and Conv1DT reset() methods
• different state_ptr iteration
• slightly more efficient handling of the variables in the forward methods
• size of layer weights is no longer dependent on dilation_rate

[jatinchowdhury18]

Nice, these changes are working great on my end.

For the formatting failure in CI, that should be easy to fix if you're able to run clang-format locally with this command.

Also, don't forget to add your name to the Contributors list!

[codecov-commenter]

Codecov Report

Merging #74 (cb4666b) into main (43834b9) will increase coverage by 0.01%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main      #74      +/-   ##
==========================================
+ Coverage   96.70%   96.72%   +0.01%     
==========================================
  Files          29       29              
  Lines        2306     2318      +12     
==========================================
+ Hits         2230     2242      +12     
  Misses         76       76              

Impacted Files	Coverage Δ
RTNeural/conv1d/conv1d_eigen.h	93.33% <100.00%> (+2.42%)	:arrow_up:
RTNeural/conv1d/conv1d_eigen.tpp	100.00% <100.00%> (ø)

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[jatinchowdhury18]

Thanks again for your great work on this!