Bugfix in c defs

[freddieknets]

Description

Fixed a small bug in cdefs before compiling a CPU kernel.
The code would check if a cdef for this kernel already exists, however, it would do so by checking by the Python name instead of the name in C.

Furthermore, I added a space before and a ( after the cname to be searched for, as otherwise if kernels exist which have the name of the new kernel as a part, the new one would not be generated.

E.g. if there already is a kernel definition void LocalSegment_crossing_drift(double s); then the new kernel void crossing_drift(double s1, double s2); would never be generated. Searching for f" {cname}(" avoids this problem.

All xobjects tests pass, and all xtrack tests with ContextCpu as well (did not test the other contexts as they are not affected).

Checklist

Mandatory:

• I have added tests to cover my changes
• All the tests are passing, including my new ones
• I described my changes in this PR description

Optional:

• The code I wrote follows good style practices (see PEP 8 and PEP 20).
• I have updated the docs in relation to my changes, if applicable
• I have tested also GPU contexts

[szymonlopaciuk]

Looks good to me! Definitely checking for pyname seems like a wrong thing to do.

I was wondering if we could add this check earlier when we construct cdefs but it seems it built partially with arbitrary string input, so it doesn't seem feasible.