New fs APIs make file concat and replace more efficient

[wkh237]

Consider the following scenario

If we wish to concat two files or replace content of one file by another, we have to cache the file content in JS context and then write the data to destination.

const fs = RNFetchBlob.fs
fs.readFile(FILE_SRC, 'base64')
     .then((b64data) => fs.appendFile(FILE_DEST, b64data, 'base64'))
     .then(() => {  console.log(`concated ${FILE_SRC} to ${FILE_DEST}`) })

This seems not efficient especially when the file is big, that cache will likely consume up the available memory.

Therefore, I think we should make writeStream, writeFile, and appendFile accept a file path as their input, and then we can do as follow.

const fs = RNFetchBlob.fs
fs.appendFile(FILE_DEST, FILE_SRC, 'uri')
     .then(() => {  console.log(`concated ${FILE_SRC} to ${FILE_DEST}`) })

• Add uri encoding type to writeFile
• Add uri encoding type to appendFile
• Test cases