properly processing CA HETATMs with JmolParser

PDBfile class in Jalview correctly processes CA HETATMs with no problem because the sequences are imported on a per-chain basis by parsing the structure file line-by-line. Each chain's sequence is terminated when a line with 'ter' key is encountered. Relevant ATMs & HETATMs are all listed before the termination statement ('ter'). See PDB entry - 1qcf.

On the other hand, importing sequences with the JMolParser follows a slightly different paradigm. The file is processed with Jmol and converted into its data model, subsequently the relevant attributes are transformed to Jalview's data model. Control of chain termination is kind of lost in the process of Jmol pre-processing, this sometimes results to generating sequences with unexpected residues (i.e when CA HETATMs are available and listed after the 'ter' statement).