Removing redundant type conversions for script scoring for hamming space with binary vectors

[kasundra07]

Description

Removes the unnecessary type conversion from byte[] to float[] and again float[] to byte for script score on KNN Binary vector using hamming space. This would help bring down the overall query latency.

For details, check #1827 (comment)

Related Issues

Resolves #1827

Check List

• New functionality includes testing.
• New functionality has been documented.
• N/A API changes companion pull request created.
• Commits are signed per the DCO using --signoff.
• N/A Public documentation issue/PR created.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[jmazanec15]

@kasundra07 I had a question around approach: #1827 (comment). Can you respond on this PR thread:

@kasundra07 Why not create a byte implementation that mirrors:

@RequiredArgsConstructor(access = AccessLevel.PRIVATE)
public abstract class KNNVectorScriptDocValues extends ScriptDocValues<float[]> {

Something like

@RequiredArgsConstructor(access = AccessLevel.PRIVATE)
public abstract class KNNByteVectorScriptDocValues extends ScriptDocValues<byte[]> {

[Vikasht34]

Wil Review this PR once this changes are completed to create the new Class for Byte.

public abstract class KNNByteVectorScriptDocValues extends ScriptDocValues<byte[]> {

[jmazanec15]

This looks good to me. Please address @VijayanB feedback