lower/upper case typos are not reported by parameter validation

[KevinMenden]

Due to the code we added to deal with camelCase/camel-case issues, parameters that are mistyped with wrong upper/lower case characters are not reported as unexpected. They are also not used by the pipeline though, so very easy to miss, when not checking the parameter summary.

Example from nf-core/smrnaseq:

When using --clip_R1 instead of --clip_r1, no warning is issued but the --clip_r1 parameter is not overwritten.

[KevinMenden]

closed by #1088